-
Notifications
You must be signed in to change notification settings - Fork 1
/
2022-08-CausalModelSelection.html
1322 lines (1147 loc) · 237 KB
/
2022-08-CausalModelSelection.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<HTML lang = "en">
<HEAD>
<meta charset="UTF-8"/>
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
tex2jax: {inlineMath: [['$','$'], ['\\(','\\)']]},
TeX: { equationNumbers: { autoNumber: "AMS" } }
});
</script>
<script type="text/javascript" async src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js?config=TeX-AMS-MML_HTMLorMML">
</script>
<style>
pre.hljl {
border: 1px solid #ccc;
margin: 5px;
padding: 5px;
overflow-x: auto;
color: rgb(68,68,68); background-color: rgb(251,251,251); }
pre.hljl > span.hljl-t { }
pre.hljl > span.hljl-w { }
pre.hljl > span.hljl-e { }
pre.hljl > span.hljl-eB { }
pre.hljl > span.hljl-o { }
pre.hljl > span.hljl-k { color: rgb(148,91,176); font-weight: bold; }
pre.hljl > span.hljl-kc { color: rgb(59,151,46); font-style: italic; }
pre.hljl > span.hljl-kd { color: rgb(214,102,97); font-style: italic; }
pre.hljl > span.hljl-kn { color: rgb(148,91,176); font-weight: bold; }
pre.hljl > span.hljl-kp { color: rgb(148,91,176); font-weight: bold; }
pre.hljl > span.hljl-kr { color: rgb(148,91,176); font-weight: bold; }
pre.hljl > span.hljl-kt { color: rgb(148,91,176); font-weight: bold; }
pre.hljl > span.hljl-n { }
pre.hljl > span.hljl-na { }
pre.hljl > span.hljl-nb { }
pre.hljl > span.hljl-nbp { }
pre.hljl > span.hljl-nc { }
pre.hljl > span.hljl-ncB { }
pre.hljl > span.hljl-nd { color: rgb(214,102,97); }
pre.hljl > span.hljl-ne { }
pre.hljl > span.hljl-neB { }
pre.hljl > span.hljl-nf { color: rgb(66,102,213); }
pre.hljl > span.hljl-nfm { color: rgb(66,102,213); }
pre.hljl > span.hljl-np { }
pre.hljl > span.hljl-nl { }
pre.hljl > span.hljl-nn { }
pre.hljl > span.hljl-no { }
pre.hljl > span.hljl-nt { }
pre.hljl > span.hljl-nv { }
pre.hljl > span.hljl-nvc { }
pre.hljl > span.hljl-nvg { }
pre.hljl > span.hljl-nvi { }
pre.hljl > span.hljl-nvm { }
pre.hljl > span.hljl-l { }
pre.hljl > span.hljl-ld { color: rgb(148,91,176); font-style: italic; }
pre.hljl > span.hljl-s { color: rgb(201,61,57); }
pre.hljl > span.hljl-sa { color: rgb(201,61,57); }
pre.hljl > span.hljl-sb { color: rgb(201,61,57); }
pre.hljl > span.hljl-sc { color: rgb(201,61,57); }
pre.hljl > span.hljl-sd { color: rgb(201,61,57); }
pre.hljl > span.hljl-sdB { color: rgb(201,61,57); }
pre.hljl > span.hljl-sdC { color: rgb(201,61,57); }
pre.hljl > span.hljl-se { color: rgb(59,151,46); }
pre.hljl > span.hljl-sh { color: rgb(201,61,57); }
pre.hljl > span.hljl-si { }
pre.hljl > span.hljl-so { color: rgb(201,61,57); }
pre.hljl > span.hljl-sr { color: rgb(201,61,57); }
pre.hljl > span.hljl-ss { color: rgb(201,61,57); }
pre.hljl > span.hljl-ssB { color: rgb(201,61,57); }
pre.hljl > span.hljl-nB { color: rgb(59,151,46); }
pre.hljl > span.hljl-nbB { color: rgb(59,151,46); }
pre.hljl > span.hljl-nfB { color: rgb(59,151,46); }
pre.hljl > span.hljl-nh { color: rgb(59,151,46); }
pre.hljl > span.hljl-ni { color: rgb(59,151,46); }
pre.hljl > span.hljl-nil { color: rgb(59,151,46); }
pre.hljl > span.hljl-noB { color: rgb(59,151,46); }
pre.hljl > span.hljl-oB { color: rgb(102,102,102); font-weight: bold; }
pre.hljl > span.hljl-ow { color: rgb(102,102,102); font-weight: bold; }
pre.hljl > span.hljl-p { }
pre.hljl > span.hljl-c { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-ch { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-cm { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-cp { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-cpB { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-cs { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-csB { color: rgb(153,153,119); font-style: italic; }
pre.hljl > span.hljl-g { }
pre.hljl > span.hljl-gd { }
pre.hljl > span.hljl-ge { }
pre.hljl > span.hljl-geB { }
pre.hljl > span.hljl-gh { }
pre.hljl > span.hljl-gi { }
pre.hljl > span.hljl-go { }
pre.hljl > span.hljl-gp { }
pre.hljl > span.hljl-gs { }
pre.hljl > span.hljl-gsB { }
pre.hljl > span.hljl-gt { }
</style>
<style type="text/css">
@font-face {
font-style: normal;
font-weight: 300;
}
@font-face {
font-style: normal;
font-weight: 400;
}
@font-face {
font-style: normal;
font-weight: 600;
}
html {
font-family: sans-serif; /* 1 */
-ms-text-size-adjust: 100%; /* 2 */
-webkit-text-size-adjust: 100%; /* 2 */
}
body {
margin: 0;
}
article,
aside,
details,
figcaption,
figure,
footer,
header,
hgroup,
main,
menu,
nav,
section,
summary {
display: block;
}
audio,
canvas,
progress,
video {
display: inline-block; /* 1 */
vertical-align: baseline; /* 2 */
}
audio:not([controls]) {
display: none;
height: 0;
}
[hidden],
template {
display: none;
}
a:active,
a:hover {
outline: 0;
}
abbr[title] {
border-bottom: 1px dotted;
}
b,
strong {
font-weight: bold;
}
dfn {
font-style: italic;
}
h1 {
font-size: 2em;
margin: 0.67em 0;
}
mark {
background: #ff0;
color: #000;
}
small {
font-size: 80%;
}
sub,
sup {
font-size: 75%;
line-height: 0;
position: relative;
vertical-align: baseline;
}
sup {
top: -0.5em;
}
sub {
bottom: -0.25em;
}
img {
border: 0;
}
svg:not(:root) {
overflow: hidden;
}
figure {
margin: 1em 40px;
}
hr {
-moz-box-sizing: content-box;
box-sizing: content-box;
height: 0;
}
pre {
overflow: auto;
}
code,
kbd,
pre,
samp {
font-family: monospace, monospace;
font-size: 1em;
}
button,
input,
optgroup,
select,
textarea {
color: inherit; /* 1 */
font: inherit; /* 2 */
margin: 0; /* 3 */
}
button {
overflow: visible;
}
button,
select {
text-transform: none;
}
button,
html input[type="button"], /* 1 */
input[type="reset"],
input[type="submit"] {
-webkit-appearance: button; /* 2 */
cursor: pointer; /* 3 */
}
button[disabled],
html input[disabled] {
cursor: default;
}
button::-moz-focus-inner,
input::-moz-focus-inner {
border: 0;
padding: 0;
}
input {
line-height: normal;
}
input[type="checkbox"],
input[type="radio"] {
box-sizing: border-box; /* 1 */
padding: 0; /* 2 */
}
input[type="number"]::-webkit-inner-spin-button,
input[type="number"]::-webkit-outer-spin-button {
height: auto;
}
input[type="search"] {
-webkit-appearance: textfield; /* 1 */
-moz-box-sizing: content-box;
-webkit-box-sizing: content-box; /* 2 */
box-sizing: content-box;
}
input[type="search"]::-webkit-search-cancel-button,
input[type="search"]::-webkit-search-decoration {
-webkit-appearance: none;
}
fieldset {
border: 1px solid #c0c0c0;
margin: 0 2px;
padding: 0.35em 0.625em 0.75em;
}
legend {
border: 0; /* 1 */
padding: 0; /* 2 */
}
textarea {
overflow: auto;
}
optgroup {
font-weight: bold;
}
table {
font-family: monospace, monospace;
font-size : 0.8em;
border-collapse: collapse;
border-spacing: 0;
}
td,
th {
padding: 0;
}
thead th {
border-bottom: 1px solid black;
background-color: white;
}
tr:nth-child(odd){
background-color: rgb(248,248,248);
}
/*
* Skeleton V2.0.4
* Copyright 2014, Dave Gamache
* www.getskeleton.com
* Free to use under the MIT license.
* http://www.opensource.org/licenses/mit-license.php
* 12/29/2014
*/
.container {
position: relative;
width: 100%;
max-width: 960px;
margin: 0 auto;
padding: 0 20px;
box-sizing: border-box; }
.column,
.columns {
width: 100%;
float: left;
box-sizing: border-box; }
@media (min-width: 400px) {
.container {
width: 85%;
padding: 0; }
}
@media (min-width: 550px) {
.container {
width: 80%; }
.column,
.columns {
margin-left: 4%; }
.column:first-child,
.columns:first-child {
margin-left: 0; }
.one.column,
.one.columns { width: 4.66666666667%; }
.two.columns { width: 13.3333333333%; }
.three.columns { width: 22%; }
.four.columns { width: 30.6666666667%; }
.five.columns { width: 39.3333333333%; }
.six.columns { width: 48%; }
.seven.columns { width: 56.6666666667%; }
.eight.columns { width: 65.3333333333%; }
.nine.columns { width: 74.0%; }
.ten.columns { width: 82.6666666667%; }
.eleven.columns { width: 91.3333333333%; }
.twelve.columns { width: 100%; margin-left: 0; }
.one-third.column { width: 30.6666666667%; }
.two-thirds.column { width: 65.3333333333%; }
.one-half.column { width: 48%; }
/* Offsets */
.offset-by-one.column,
.offset-by-one.columns { margin-left: 8.66666666667%; }
.offset-by-two.column,
.offset-by-two.columns { margin-left: 17.3333333333%; }
.offset-by-three.column,
.offset-by-three.columns { margin-left: 26%; }
.offset-by-four.column,
.offset-by-four.columns { margin-left: 34.6666666667%; }
.offset-by-five.column,
.offset-by-five.columns { margin-left: 43.3333333333%; }
.offset-by-six.column,
.offset-by-six.columns { margin-left: 52%; }
.offset-by-seven.column,
.offset-by-seven.columns { margin-left: 60.6666666667%; }
.offset-by-eight.column,
.offset-by-eight.columns { margin-left: 69.3333333333%; }
.offset-by-nine.column,
.offset-by-nine.columns { margin-left: 78.0%; }
.offset-by-ten.column,
.offset-by-ten.columns { margin-left: 86.6666666667%; }
.offset-by-eleven.column,
.offset-by-eleven.columns { margin-left: 95.3333333333%; }
.offset-by-one-third.column,
.offset-by-one-third.columns { margin-left: 34.6666666667%; }
.offset-by-two-thirds.column,
.offset-by-two-thirds.columns { margin-left: 69.3333333333%; }
.offset-by-one-half.column,
.offset-by-one-half.columns { margin-left: 52%; }
}
html {
font-size: 62.5%; }
body {
font-size: 1.5em; /* currently ems cause chrome bug misinterpreting rems on body element */
line-height: 1.6;
font-weight: 400;
font-family: "Raleway", "HelveticaNeue", "Helvetica Neue", Helvetica, Arial, sans-serif;
color: #222; }
h1, h2, h3, h4, h5, h6 {
margin-top: 0;
margin-bottom: 2rem;
font-weight: 300; }
h1 { font-size: 3.6rem; line-height: 1.2; letter-spacing: -.1rem;}
h2 { font-size: 3.4rem; line-height: 1.25; letter-spacing: -.1rem; }
h3 { font-size: 3.2rem; line-height: 1.3; letter-spacing: -.1rem; }
h4 { font-size: 2.8rem; line-height: 1.35; letter-spacing: -.08rem; }
h5 { font-size: 2.4rem; line-height: 1.5; letter-spacing: -.05rem; }
h6 { font-size: 1.5rem; line-height: 1.6; letter-spacing: 0; }
p {
margin-top: 0; }
a {
color: #1EAEDB; }
a:hover {
color: #0FA0CE; }
.button,
button,
input[type="submit"],
input[type="reset"],
input[type="button"] {
display: inline-block;
height: 38px;
padding: 0 30px;
color: #555;
text-align: center;
font-size: 11px;
font-weight: 600;
line-height: 38px;
letter-spacing: .1rem;
text-transform: uppercase;
text-decoration: none;
white-space: nowrap;
background-color: transparent;
border-radius: 4px;
border: 1px solid #bbb;
cursor: pointer;
box-sizing: border-box; }
.button:hover,
button:hover,
input[type="submit"]:hover,
input[type="reset"]:hover,
input[type="button"]:hover,
.button:focus,
button:focus,
input[type="submit"]:focus,
input[type="reset"]:focus,
input[type="button"]:focus {
color: #333;
border-color: #888;
outline: 0; }
.button.button-primary,
button.button-primary,
input[type="submit"].button-primary,
input[type="reset"].button-primary,
input[type="button"].button-primary {
color: #FFF;
background-color: #33C3F0;
border-color: #33C3F0; }
.button.button-primary:hover,
button.button-primary:hover,
input[type="submit"].button-primary:hover,
input[type="reset"].button-primary:hover,
input[type="button"].button-primary:hover,
.button.button-primary:focus,
button.button-primary:focus,
input[type="submit"].button-primary:focus,
input[type="reset"].button-primary:focus,
input[type="button"].button-primary:focus {
color: #FFF;
background-color: #1EAEDB;
border-color: #1EAEDB; }
input[type="email"],
input[type="number"],
input[type="search"],
input[type="text"],
input[type="tel"],
input[type="url"],
input[type="password"],
textarea,
select {
height: 38px;
padding: 6px 10px; /* The 6px vertically centers text on FF, ignored by Webkit */
background-color: #fff;
border: 1px solid #D1D1D1;
border-radius: 4px;
box-shadow: none;
box-sizing: border-box; }
/* Removes awkward default styles on some inputs for iOS */
input[type="email"],
input[type="number"],
input[type="search"],
input[type="text"],
input[type="tel"],
input[type="url"],
input[type="password"],
textarea {
-webkit-appearance: none;
-moz-appearance: none;
appearance: none; }
textarea {
min-height: 65px;
padding-top: 6px;
padding-bottom: 6px; }
input[type="email"]:focus,
input[type="number"]:focus,
input[type="search"]:focus,
input[type="text"]:focus,
input[type="tel"]:focus,
input[type="url"]:focus,
input[type="password"]:focus,
textarea:focus,
select:focus {
border: 1px solid #33C3F0;
outline: 0; }
label,
legend {
display: block;
margin-bottom: .5rem;
font-weight: 600; }
fieldset {
padding: 0;
border-width: 0; }
input[type="checkbox"],
input[type="radio"] {
display: inline; }
label > .label-body {
display: inline-block;
margin-left: .5rem;
font-weight: normal; }
ul {
list-style: circle; }
ol {
list-style: decimal; }
ul ul,
ul ol,
ol ol,
ol ul {
margin: 1.5rem 0 1.5rem 3rem;
font-size: 90%; }
li > p {margin : 0;}
th,
td {
padding: 12px 15px;
text-align: left;
border-bottom: 1px solid #E1E1E1; }
th:first-child,
td:first-child {
padding-left: 0; }
th:last-child,
td:last-child {
padding-right: 0; }
button,
.button {
margin-bottom: 1rem; }
input,
textarea,
select,
fieldset {
margin-bottom: 1.5rem; }
pre,
blockquote,
dl,
figure,
table,
p,
ul,
ol,
form {
margin-bottom: 1.0rem; }
.u-full-width {
width: 100%;
box-sizing: border-box; }
.u-max-full-width {
max-width: 100%;
box-sizing: border-box; }
.u-pull-right {
float: right; }
.u-pull-left {
float: left; }
hr {
margin-top: 3rem;
margin-bottom: 3.5rem;
border-width: 0;
border-top: 1px solid #E1E1E1; }
.container:after,
.row:after,
.u-cf {
content: "";
display: table;
clear: both; }
pre {
display: block;
padding: 9.5px;
margin: 0 0 10px;
font-size: 13px;
line-height: 1.42857143;
word-break: break-all;
word-wrap: break-word;
border: 1px solid #ccc;
border-radius: 4px;
}
pre.hljl {
margin: 0 0 10px;
display: block;
background: #f5f5f5;
border-radius: 4px;
padding : 5px;
}
pre.output {
background: #ffffff;
}
pre.code {
background: #ffffff;
}
pre.julia-error {
color : red
}
code,
kbd,
pre,
samp {
font-family: Menlo, Monaco, Consolas, "Courier New", monospace;
font-size: 0.9em;
}
@media (min-width: 400px) {}
@media (min-width: 550px) {}
@media (min-width: 750px) {}
@media (min-width: 1000px) {}
@media (min-width: 1200px) {}
h1.title {margin-top : 20px}
img {max-width : 100%}
div.title {text-align: center;}
</style>
</HEAD>
<BODY>
<div class ="container">
<div class = "row">
<div class = "col-md-12 twelve columns">
<div class="title">
</div>
<h1>Causal model selection tutorial</h1>
<h4>Background</h4>
<p>In this tutorial we consider two continuous and correlated variables, <span class="math">$X$</span> and <span class="math">$Y$</span>, representing the expression levels of two genes. We also consider a discrete variable <span class="math">$Z$</span> representing a genotype for gene <span class="math">$X$</span>. Typically, <span class="math">$Z$</span> will have been obtained by <a href="https://en.wikipedia.org/wiki/Expression_quantitative_trait_loci">eQTL mapping</a> for gene <span class="math">$X$</span>. We wish to determine if variation in <span class="math">$X$</span> causes variation in <span class="math">$Y$</span>.</p>
<p>The <strong>aim of causal model selection</strong> is: given observational data for <span class="math">$Z$</span>, <span class="math">$X$</span>, and <span class="math">$Y$</span> in a set of independent samples, which causal model (represented by a directed acyclic graph) explains the data best. </p>
<p>To restrict the space of possible models that need to be considered, a number of assumptions reflecting biological knowledge can be made:</p>
<ol>
<li><p>Genetic variation influences variation in gene expression (and phenotypes more generally), but changing the value of an individual's phenotype does not change their genome. Hence, in our models there can be <strong>no incoming arrows into <span class="math">$Z$</span></strong>.</p>
</li>
<li><p>We assume that the statistical association between <span class="math">$Z$</span> and <span class="math">$X$</span> is due to a <em>direct</em> effect, that is, all causal models we consider <strong>must contain the directed edge <span class="math">$Z\to X$</span></strong>. This assumption is justified if <span class="math">$Z$</span> is located within or near to <span class="math">$X$</span> (on the genome) and in a known regulatory region for <span class="math">$X$</span>.</p>
</li>
<li><p>For <span class="math">$X$</span> and <span class="math">$Y$</span> to be correlated (non-independent), there must be a path in the graph between them, more precisely <strong><span class="math">$X$</span> and <span class="math">$Y$</span> must not be</strong> <a href="http://bayes.cs.ucla.edu/BOOK-2K/d-sep.html"><strong>d-separated</strong></a>.</p>
</li>
<li><p>By the laws of <a href="https://en.wikipedia.org/wiki/Mendelian_inheritance">Mendelian inheritance</a> (particularly the random segregation of alleles), we may assume that <span class="math">$Z$</span> is independent of any unobserved confounding factors <span class="math">$U$</span> that cause <span class="math">$X$</span> and <span class="math">$Y$</span> to be correlated, and therefore <strong>there are no edges between <span class="math">$Z$</span> and any unobserved <span class="math">$U$</span></strong>.</p>
</li>
</ol>
<p>If we assume that there are <strong>no unobserved factors</strong>, there are 5 possible models satisfying assumptions 1-4 (see figure below). If we allow for the <strong>presence of unobserved factors</strong>, we have the same 5 models with an additional unobserved <span class="math">$U$</span> having a causal effect on <span class="math">$X$</span> and <span class="math">$Y$</span>, plus one more model without an edge between <span class="math">$X$</span> and <span class="math">$Y$</span> where all of their correlation is explained by <span class="math">$U$</span> (see figure).</p>
<p>Below each model in the figure, some of the key conditional independences implied by the model are shown, using the mathematical notation <span class="math">$⫫$</span> for "is independent of", <span class="math">$∣$</span> for "conditional on", <span class="math">$∧$</span> for "and", and <span class="math">$¬$</span> for "it is not the case that".</p>
<p><img src="./figures/causal_models.png" alt="Causal models" /></p>
<p>From this figure it is immediately clear that our aim of deciding whether <span class="math">$X$</span> causes <span class="math">$Y$</span> is unachievable. Even without any unobserved factors, models 4a and 5a are <a href="https://www.cs.ru.nl/~peterl/markoveq.pdf">Markov equivalent</a>: they entail the same conditional independences and are indistinguishable using observational data. In other words, <strong>there exists no condition that is both necessary and sufficient</strong> for <span class="math">$X$</span> to cause <span class="math">$Y$</span> given the above 4 assumptions.</p>
<p>There are two possible paths forward (apart from giving up): to <strong>use a sufficient condition or a necessary condition</strong> for causality. If the joint probability distribution of <span class="math">$Z$</span>, <span class="math">$X$</span>, and <span class="math">$Y$</span> passes a sufficient condition, then it is guaranteed to have been generated by a model where <span class="math">$X$</span> causes <span class="math">$Y$</span>, but there may be distributions with <span class="math">$X\to Y$</span> that don't pass the test (false negatives). Conversely, all joint distributions generated by models with <span class="math">$X\to Y$</span> will pass a necessary condition, but there may be distributions generated by models where <span class="math">$X$</span> does not cause <span class="math">$Y$</span> that also pass the test (false positives).</p>
<p>Because estimating the joint distribution of <span class="math">$Z$</span>, <span class="math">$X$</span>, and <span class="math">$Y$</span> and its conditional independences from a finite number of samples is itself an imperfect process that can only ever approximate the true distribution, having additional errors from using an imperfect causality test is not necessarily catastrophic, provided those errors are small enough. </p>
<h4>A sufficient condition for <span class="math">$X\to Y$</span></h4>
<p>Our sufficient condition for <span class="math">$X\to Y$</span> is based on model 1a. This model implies that <span class="math">$Z$</span> and <span class="math">$Y$</span> are not independent, but <span class="math">$Z$</span> is independent of <span class="math">$Y$</span> conditioned on <span class="math">$X$</span> (<span class="math">$X$</span> is a <strong>mediator</strong> for the causal path from <span class="math">$Z$</span> to <span class="math">$Y$</span>). No other model satisfies those two relations:</p>
<ul>
<li><p>In models 2a, 2b, and 6b, <span class="math">$Z$</span> and <span class="math">$Y$</span> are independent.</p>
</li>
<li><p>In all other models <span class="math">$Z$</span> is not independent of <span class="math">$Y$</span> conditioned on <span class="math">$X$</span>, either because there is a direct path from <span class="math">$Z$</span> to <span class="math">$Y$</span> not passing through <span class="math">$X$</span>, or because conditioning on <span class="math">$X$</span> opens a path from <span class="math">$Z$</span> to <span class="math">$Y$</span> via the confounder <span class="math">$U$</span> (due to the v-structure or <a href="https://en.wikipedia.org/wiki/Collider_(statistics)">collider</a> at <span class="math">$X$</span>).</p>
</li>
</ul>
<p>Hence any joint distribution in which <span class="math">$Z$</span> and <span class="math">$Y$</span> are not independent, but <span class="math">$Z$</span> is independent of <span class="math">$Y$</span> conditioned on <span class="math">$X$</span>, must have been generated by model 1a, that is, by a model where <span class="math">$X\to Y$</span>. In mathematical notation:</p>
<p class="math">\[
¬(Z⫫Y) ∧ (Z⫫Y∣X) ⇒ (X→Y)
\]</p>
<h4>A necessary condition for <span class="math">$X\to Y$</span></h4>
<p>Because all models contain the edge <span class="math">$G→X$</span>, it follows that if <span class="math">$X→Y$</span>, then <span class="math">$Z$</span> and <span class="math">$Y$</span> cannot be independent, providing a simple necessary condition for <span class="math">$X→Y$</span>. However, <span class="math">$Z$</span> and <span class="math">$Y$</span> are also not independent in models 3a-b and 5a-b, in which <span class="math">$Y→X$</span>, because of the direct edge <span class="math">$G→Y$</span>. Of these, only 3a can be excluded, because in 3a, <span class="math">$X⫫Y∣Z$</span>, a condition not satisfied in any model where <span class="math">$X→Y$</span>. Combining these two results, we obtain</p>
<p class="math">\[
(X→Y) ⇒ ¬(Z⫫Y) ∧ ¬(X⫫Y∣Z)
\]</p>
<h2>Import packages</h2>
<p>In this tutorial, we will first set up some code to generate and visualize simulated data for <span class="math">$Z$</span>, <span class="math">$X$</span>, and <span class="math">$Y$</span> given one of the models in the figure above. Then we will implement the sufficient and necessary conditions for <span class="math">$X→Y$</span> and assess how well they perform in a number of scenarios. The code is written in <a href="https://julialang.org">julia</a>. It should be fairly straightforward to translate into other languages.</p>
<p>We start by importing some necessary packages:</p>
<pre class='hljl'>
<span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>Random</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>Distributions</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>StatsBase</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>DataFrames</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>GLM</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>LinearAlgebra</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>CairoMakie</span><span class='hljl-t'>
</span><span class='hljl-k'>using</span><span class='hljl-t'> </span><span class='hljl-n'>LaTeXStrings</span>
</pre>
<h2>Data simulation</h2>
<p>First we set a random seed and fix the number of samples:</p>
<pre class='hljl'>
<span class='hljl-n'>Random</span><span class='hljl-oB'>.</span><span class='hljl-nf'>seed!</span><span class='hljl-p'>(</span><span class='hljl-ni'>123</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-n'>N</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-ni'>200</span><span class='hljl-p'>;</span>
</pre>
<p>In our simulations, as in real biology, a genotype value is sampled first. Then expression values for the two genes are sampled conditional on the genotype value.</p>
<h3>Genotype data simulation</h3>
<p>We simulate the genotype of a bi-allelic (2 values), diploid (2 copies) polymorphism. We encode the major and minor alleles by the values 0 and 1, respectively. The genotype is sampled by defining the minor allele frequency (MAF) as a parameter of the simulation. Two haploids are sampled using a Bernoulli distribution and summed to give the genotype, that is, the genotype is the number of copies of the minor allele in an individual. </p>
<p>Because the mean of a Bernoulli distribution with probability of success <span class="math">$p$</span> is <span class="math">$p$</span>, the mean of <span class="math">$Z$</span> is 2 times the minor allele frequency, and we therefore subtract this value from the sampled genotypes to obtain samples from a random variable <span class="math">$Z$</span> with mean zero. Note that we <em>cannot</em> center the actual sampled values at this point, because we still need to generate the expression data conditional on <span class="math">$Z$</span>, and the expression levels of an individual cannot depend on the sample mean of the genotypes in a population! </p>
<pre class='hljl'>
<span class='hljl-n'>maf</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nfB'>0.3</span><span class='hljl-t'>
</span><span class='hljl-n'>H1</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>rand</span><span class='hljl-p'>(</span><span class='hljl-nf'>Bernoulli</span><span class='hljl-p'>(</span><span class='hljl-n'>maf</span><span class='hljl-p'>),</span><span class='hljl-n'>N</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-n'>H2</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>rand</span><span class='hljl-p'>(</span><span class='hljl-nf'>Bernoulli</span><span class='hljl-p'>(</span><span class='hljl-n'>maf</span><span class='hljl-p'>),</span><span class='hljl-n'>N</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-n'>Z0</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-n'>H1</span><span class='hljl-t'> </span><span class='hljl-oB'>.+</span><span class='hljl-t'> </span><span class='hljl-n'>H2</span><span class='hljl-t'>
</span><span class='hljl-n'>Z</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-n'>Z0</span><span class='hljl-t'> </span><span class='hljl-oB'>.-</span><span class='hljl-t'> </span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>maf</span><span class='hljl-p'>;</span>
</pre>
<h3>Expression data simulation</h3>
<p>To simulate data for <span class="math">$X$</span> and <span class="math">$Y$</span>, we must first set up the <a href="https://en.wikipedia.org/wiki/Structural_equation_modeling">structural equations</a> for the <a href="https://en.wikipedia.org/wiki/Causal_model">causal model</a> we want to simulate. We will assume linear models with additive Gaussian noise throughout. </p>
<p>Let's start by models 1a, 1b, and 6b. Their structural equations can be written as </p>
<p class="math">\[
\begin{align}
X &= a Z + U_X\\
Y &= b X + U_Y
\end{align}
\]</p>
<p>where <span class="math">$U_X$</span> and <span class="math">$U_Y$</span> are normally distributed errors with joint distribution</p>
<p class="math">\[
(U_X, U_Y)^T \sim {\cal N}(0,Σ_{U})
\]</p>
<p>with covariance matrix </p>
<p class="math">\[
Σ_{U} =
\begin{pmatrix}
1 & ρ\\
ρ & 1
\end{pmatrix}
\]</p>
<p>In model 1a, the errors are uncorrelated, <span class="math">$\rho=0$</span>, and we arbitrarily set the errors to have unit variance. In model 1b, the unobserved confounder <span class="math">$U$</span> has the effect of correlating the errors, that is <span class="math">$0<\rho<1$</span>. In model 6b, the errors are also correlated, <span class="math">$0<\rho<1$</span>, but there is no direct effect of <span class="math">$X$</span> on <span class="math">$Y$</span>, that is <span class="math">$b=0$</span>.</p>
<p>The parameters <span class="math">$a$</span> and <span class="math">$b$</span> are the causal effect sizes, and their magnitudes should be interpreted relative to the unit standard deviation of the random errors. In other words, each additional alternative allele shifts the mean of <span class="math">$X$</span> by <span class="math">$a$</span> standard deviations of the random errors.</p>
<p>Given a value <span class="math">$Z=z$</span>, eqs. (1)–(2) can be rewritten in matrix-vector notation as</p>
<p class="math">\[
\begin{pmatrix}
X\\
Y
\end{pmatrix}
=
\begin{pmatrix}
az \\
abz
\end{pmatrix} +
\begin{pmatrix}
1 & 0\\
b & 1
\end{pmatrix}
\begin{pmatrix}
U_X\\
U_Y
\end{pmatrix}
\]</p>
<p>Using <a href="https://en.wikipedia.org/wiki/Multivariate_normal_distribution#Affine_transformation">properties of the multivariate normal distribution</a>, it follows that </p>
<p class="math">\[
\begin{equation}
\begin{pmatrix}
X\\
Y
\end{pmatrix} \mid Z=z
∼ {\cal N} \left( \mu_z, Σ_{XY} \right)
\end{equation}
\]</p>
<p>with</p>
<p class="math">\[
\begin{align*}
\mu_z &= \begin{pmatrix}
az \\
abz
\end{pmatrix} \\
Σ_{XY} &=
\begin{pmatrix}
1 & 0\\
b & 1
\end{pmatrix}
\begin{pmatrix}
1 & ρ\\
ρ & 1
\end{pmatrix}
\begin{pmatrix}
1 & b \\
0 & 1
\end{pmatrix}
= \begin{pmatrix}
1 & b+ρ\\
b+ρ & b^2 + 2b\rho + 1
\end{pmatrix}
\end{align*}
\]</p>
<p>Hence given a sampled genotype value <span class="math">$Z$</span>, we can sample values for <span class="math">$X$</span> and <span class="math">$Y$</span> by sampling from the multivariate normal distribution (3). In fact, <a href="https://julialang.org/">julia</a> knows how to do calculus with <a href="https://juliastats.org/Distributions.jl/stable/">distributions</a>, and we only need to specify the distribution of the random errors and the affine matrix-vector transformation:</p>
<pre class='hljl'>
<span class='hljl-k'>function</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_XcauseY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-n'>a</span><span class='hljl-oB'>=</span><span class='hljl-nfB'>1.0</span><span class='hljl-p'>,</span><span class='hljl-n'>b</span><span class='hljl-oB'>=</span><span class='hljl-nfB'>0.8</span><span class='hljl-p'>,</span><span class='hljl-n'>rho</span><span class='hljl-oB'>=</span><span class='hljl-nfB'>0.0</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-cs'># number of samples</span><span class='hljl-t'>
</span><span class='hljl-n'>N</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>length</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-cs'># Covariance matrix for the unmodelled factors</span><span class='hljl-t'>
</span><span class='hljl-n'>covU</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-nfB'>1.0</span><span class='hljl-t'> </span><span class='hljl-n'>rho</span><span class='hljl-t'>
</span><span class='hljl-n'>rho</span><span class='hljl-t'> </span><span class='hljl-nfB'>1.0</span><span class='hljl-p'>]</span><span class='hljl-t'>
</span><span class='hljl-cs'># distribution of the unmodelled factors</span><span class='hljl-t'>
</span><span class='hljl-n'>distU</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>MvNormal</span><span class='hljl-p'>(</span><span class='hljl-n'>covU</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-cs'># the covariance between X and Y due to direct and unmodelled effects</span><span class='hljl-t'>
</span><span class='hljl-n'>B</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-nfB'>1.</span><span class='hljl-t'> </span><span class='hljl-nfB'>0.</span><span class='hljl-t'>
</span><span class='hljl-n'>b</span><span class='hljl-t'> </span><span class='hljl-nfB'>1.</span><span class='hljl-p'>]</span><span class='hljl-t'>
</span><span class='hljl-n'>distXY</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-n'>B</span><span class='hljl-t'> </span><span class='hljl-oB'>*</span><span class='hljl-t'> </span><span class='hljl-n'>distU</span><span class='hljl-t'>
</span><span class='hljl-cs'># sample XY expression levels from model 1a, 1b, or 6b, depending on the values of rho and b</span><span class='hljl-t'>
</span><span class='hljl-n'>XY</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-n'>a</span><span class='hljl-oB'>*</span><span class='hljl-n'>Z</span><span class='hljl-t'> </span><span class='hljl-n'>a</span><span class='hljl-oB'>*</span><span class='hljl-n'>b</span><span class='hljl-oB'>*</span><span class='hljl-n'>Z</span><span class='hljl-p'>]</span><span class='hljl-t'> </span><span class='hljl-oB'>+</span><span class='hljl-t'> </span><span class='hljl-nf'>rand</span><span class='hljl-p'>(</span><span class='hljl-n'>distXY</span><span class='hljl-p'>,</span><span class='hljl-n'>N</span><span class='hljl-p'>)</span><span class='hljl-oB'>'</span><span class='hljl-t'>
</span><span class='hljl-k'>return</span><span class='hljl-t'> </span><span class='hljl-n'>XY</span><span class='hljl-t'>
</span><span class='hljl-k'>end</span><span class='hljl-p'>;</span>
</pre>
<p><strong>Your turn:</strong> Can you derive and implement the structural equations for the other models? </p>
<p>The function to sample from models 3a or 3b is as follows:</p>
<pre class='hljl'>
<span class='hljl-k'>function</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_GcauseXY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-n'>a</span><span class='hljl-oB'>=</span><span class='hljl-nfB'>1.0</span><span class='hljl-p'>,</span><span class='hljl-n'>b</span><span class='hljl-oB'>=</span><span class='hljl-nfB'>0.8</span><span class='hljl-p'>,</span><span class='hljl-n'>rho</span><span class='hljl-oB'>=</span><span class='hljl-nfB'>0.0</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-cs'># number of samples</span><span class='hljl-t'>
</span><span class='hljl-n'>N</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>length</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-cs'># Covariance matrix for the unmodelled factors</span><span class='hljl-t'>
</span><span class='hljl-n'>covU</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-nfB'>1.0</span><span class='hljl-t'> </span><span class='hljl-n'>rho</span><span class='hljl-t'>
</span><span class='hljl-n'>rho</span><span class='hljl-t'> </span><span class='hljl-nfB'>1.0</span><span class='hljl-p'>]</span><span class='hljl-t'>
</span><span class='hljl-cs'># distribution of unmodelled factors</span><span class='hljl-t'>
</span><span class='hljl-n'>distU</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>MvNormal</span><span class='hljl-p'>(</span><span class='hljl-n'>covU</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-cs'># sample XY expression levels from model 3a or 3b, depending on the value of rho</span><span class='hljl-t'>
</span><span class='hljl-n'>XY</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-n'>a</span><span class='hljl-oB'>*</span><span class='hljl-n'>Z</span><span class='hljl-t'> </span><span class='hljl-n'>b</span><span class='hljl-oB'>*</span><span class='hljl-n'>Z</span><span class='hljl-p'>]</span><span class='hljl-t'> </span><span class='hljl-oB'>+</span><span class='hljl-t'> </span><span class='hljl-nf'>rand</span><span class='hljl-p'>(</span><span class='hljl-n'>distU</span><span class='hljl-p'>,</span><span class='hljl-n'>N</span><span class='hljl-p'>)</span><span class='hljl-oB'>'</span><span class='hljl-t'>
</span><span class='hljl-k'>return</span><span class='hljl-t'> </span><span class='hljl-n'>XY</span><span class='hljl-t'>
</span><span class='hljl-k'>end</span><span class='hljl-p'>;</span>
</pre>
<p>Now sample some data:</p>
<pre class='hljl'>
<span class='hljl-n'>XY1a</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_XcauseY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-nfB'>1.</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.8</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.</span><span class='hljl-p'>)</span><span class='hljl-t'> </span><span class='hljl-cs'># sample from model 1a</span><span class='hljl-t'>
</span><span class='hljl-n'>XY1b</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_XcauseY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-nfB'>1.</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.8</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.4</span><span class='hljl-p'>)</span><span class='hljl-t'> </span><span class='hljl-cs'># sample from model 1b</span><span class='hljl-t'>
</span><span class='hljl-n'>XY6b</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_XcauseY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-nfB'>1.</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.4</span><span class='hljl-p'>);</span><span class='hljl-t'> </span><span class='hljl-cs'># sample from model 6b</span><span class='hljl-t'>
</span><span class='hljl-n'>XY3a</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_GcauseXY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-nfB'>1.</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.8</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.</span><span class='hljl-p'>)</span><span class='hljl-t'> </span><span class='hljl-cs'># sample from model 3a</span><span class='hljl-t'>
</span><span class='hljl-n'>XY3b</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>sample_GcauseXY</span><span class='hljl-p'>(</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-nfB'>1.</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.8</span><span class='hljl-p'>,</span><span class='hljl-nfB'>0.4</span><span class='hljl-p'>);</span><span class='hljl-t'> </span><span class='hljl-cs'># sample from model 3b</span>
</pre>
<p>Let's collect all our data in a dataframe:</p>
<pre class='hljl'>
<span class='hljl-n'>data</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>DataFrame</span><span class='hljl-p'>(</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>=</span><span class='hljl-n'>Z0</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>Z</span><span class='hljl-oB'>=</span><span class='hljl-n'>Z</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>X1a</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY1a</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-t'> </span><span class='hljl-n'>Y1a</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY1a</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>X1b</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY1b</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-t'> </span><span class='hljl-n'>Y1b</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY1b</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>X6b</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY6b</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-t'> </span><span class='hljl-n'>Y6b</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY6b</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>X3a</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY3a</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-t'> </span><span class='hljl-n'>Y3a</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY3a</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>X3b</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY3b</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-t'> </span><span class='hljl-n'>Y3b</span><span class='hljl-oB'>=</span><span class='hljl-n'>XY3b</span><span class='hljl-p'>[</span><span class='hljl-oB'>:</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-p'>]);</span>
</pre>
<p>We can visualize the data using scatter plots of <span class="math">$X$</span> and <span class="math">$Y$</span> colored by genotype value:</p>
<pre class='hljl'>
<span class='hljl-n'>f1</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>Figure</span><span class='hljl-p'>()</span><span class='hljl-t'>
</span><span class='hljl-n'>figttl</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-s'>"Model 1a"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-s'>"Model 1b"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-s'>"Model 6b"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-s'>"Model 3a"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-s'>"Model 3b"</span><span class='hljl-p'>]</span><span class='hljl-t'>
</span><span class='hljl-n'>rowid</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>2</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>2</span><span class='hljl-p'>]</span><span class='hljl-t'>
</span><span class='hljl-n'>colid</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-p'>[</span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>2</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>3</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-ni'>2</span><span class='hljl-p'>]</span><span class='hljl-t'>
</span><span class='hljl-k'>for</span><span class='hljl-t'> </span><span class='hljl-n'>k</span><span class='hljl-oB'>=</span><span class='hljl-ni'>1</span><span class='hljl-oB'>:</span><span class='hljl-ni'>5</span><span class='hljl-t'>
</span><span class='hljl-n'>ax</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>Axis</span><span class='hljl-p'>(</span><span class='hljl-n'>f1</span><span class='hljl-p'>[</span><span class='hljl-n'>rowid</span><span class='hljl-p'>[</span><span class='hljl-n'>k</span><span class='hljl-p'>],</span><span class='hljl-n'>colid</span><span class='hljl-p'>[</span><span class='hljl-n'>k</span><span class='hljl-p'>]],</span><span class='hljl-t'> </span><span class='hljl-n'>xlabel</span><span class='hljl-oB'>=</span><span class='hljl-s'>"x"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>ylabel</span><span class='hljl-oB'>=</span><span class='hljl-s'>"y"</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>title</span><span class='hljl-oB'>=</span><span class='hljl-n'>figttl</span><span class='hljl-p'>[</span><span class='hljl-n'>k</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>aspect</span><span class='hljl-oB'>=</span><span class='hljl-nf'>AxisAspect</span><span class='hljl-p'>(</span><span class='hljl-ni'>1</span><span class='hljl-p'>))</span><span class='hljl-t'>
</span><span class='hljl-nf'>hidespines!</span><span class='hljl-p'>(</span><span class='hljl-n'>ax</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-nf'>scatter!</span><span class='hljl-p'>(</span><span class='hljl-n'>data</span><span class='hljl-p'>[</span><span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>.==</span><span class='hljl-ni'>0</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>k</span><span class='hljl-oB'>+</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-n'>data</span><span class='hljl-p'>[</span><span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>.==</span><span class='hljl-ni'>0</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>k</span><span class='hljl-oB'>+</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>label</span><span class='hljl-oB'>=</span><span class='hljl-s'>"z=0"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>marker</span><span class='hljl-oB'>=:</span><span class='hljl-n'>circle</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>color</span><span class='hljl-oB'>=:</span><span class='hljl-n'>black</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>markersize</span><span class='hljl-oB'>=</span><span class='hljl-ni'>7</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-nf'>scatter!</span><span class='hljl-p'>(</span><span class='hljl-n'>data</span><span class='hljl-p'>[</span><span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>.==</span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>k</span><span class='hljl-oB'>+</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-n'>data</span><span class='hljl-p'>[</span><span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>.==</span><span class='hljl-ni'>1</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>k</span><span class='hljl-oB'>+</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>label</span><span class='hljl-oB'>=</span><span class='hljl-s'>"z=1"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>marker</span><span class='hljl-oB'>=:</span><span class='hljl-n'>xcross</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>color</span><span class='hljl-oB'>=:</span><span class='hljl-n'>lightblue</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>markersize</span><span class='hljl-oB'>=</span><span class='hljl-ni'>12</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-nf'>scatter!</span><span class='hljl-p'>(</span><span class='hljl-n'>data</span><span class='hljl-p'>[</span><span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>.==</span><span class='hljl-ni'>2</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>k</span><span class='hljl-oB'>+</span><span class='hljl-ni'>1</span><span class='hljl-p'>],</span><span class='hljl-n'>data</span><span class='hljl-p'>[</span><span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>Z0</span><span class='hljl-oB'>.==</span><span class='hljl-ni'>2</span><span class='hljl-p'>,</span><span class='hljl-ni'>2</span><span class='hljl-oB'>*</span><span class='hljl-n'>k</span><span class='hljl-oB'>+</span><span class='hljl-ni'>2</span><span class='hljl-p'>],</span><span class='hljl-t'>
</span><span class='hljl-n'>label</span><span class='hljl-oB'>=</span><span class='hljl-s'>"z=2"</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>marker</span><span class='hljl-oB'>=:</span><span class='hljl-n'>cross</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>color</span><span class='hljl-t'> </span><span class='hljl-oB'>=:</span><span class='hljl-n'>red</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>markersize</span><span class='hljl-oB'>=</span><span class='hljl-ni'>12</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-n'>f1</span><span class='hljl-p'>[</span><span class='hljl-ni'>2</span><span class='hljl-p'>,</span><span class='hljl-ni'>3</span><span class='hljl-p'>]</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>Legend</span><span class='hljl-p'>(</span><span class='hljl-n'>f1</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>ax</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-s'>"Genotype"</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>framevisible</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-kc'>false</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>tellheight</span><span class='hljl-oB'>=</span><span class='hljl-kc'>false</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>tellwidth</span><span class='hljl-oB'>=</span><span class='hljl-kc'>false</span><span class='hljl-p'>,</span><span class='hljl-t'>
</span><span class='hljl-n'>halign</span><span class='hljl-oB'>=:</span><span class='hljl-n'>left</span><span class='hljl-p'>,</span><span class='hljl-t'> </span><span class='hljl-n'>valign</span><span class='hljl-oB'>=:</span><span class='hljl-n'>bottom</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-k'>end</span><span class='hljl-t'>
</span><span class='hljl-n'>f1</span>
</pre>
<img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAyAAAAJYCAIAAAAVFBUnAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nOzdeXxU1dk48OfcZfZ9sieTsBOQfQ2QIKgIImJr/UlFbGtfbVXEKlDr+tattX0BtcW6tGqr4vbWXcuruKBA2CFAIJEtJJnJNpl9n7ud3x83GYeEhABJJiHn+/GP3Jk79x7inMwz5zznOQhjDARBEARBEET3oVLdAIIgCIIgiIsNCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsAiCIAiCILoZCbAIgiAIgiC6GQmwCIIgCIIguhkJsHrVc889hxCiabquri758f/7v/9DCCGEvv322/O47BVXXLFo0aKunPnkk0/qdLqOnpUk6euvv77lllsMBgNCqLq6+jwaQxDdq4d6DXRTx+m816xevTorK+v8mkcQyfr4x4ds48aNl156qV6vt9ls9913XzAYlB8fmB2BBFgpoFKp3nnnneRH3nzzTY1Gk6r2JOzateuPf/xjSUnJ6tWrU90WgjgN6TUEAX24IwDASy+9dOONN9522211dXUHDx7Mysr68MMPU92oVCIBVgosXrz4zTffTByGw+GPPvro2muvTWGTZDNmzPj6669/+ctfGgyGVLeFIE5Deg1BQB/uCNXV1ffcc8/zzz+/bNkyg8FgsVhWrlz5s5/9LNXtSiUSYKXA0qVLy8rKvv/+e/nw448/BoD2PeTrr78uLi7WaDRGo3Hx4sWJ8wHgvffeGzVqlFqtLioqKisra3+L8vLya6+91mw2q9XqmTNnbtmy5cKb/fLLL8sD0TRNFxQU/OxnP7Pb7Rd+WYLoigvvNXC2jtMTvUZWX19/1VVXabXagoKCZ555prsuSwxAXekIKekFr776qlKpXLJkSSfnDLSOQAKsFJg4ceKoUaMS30Leeuuta6+9VqvVJp/z9ddfz58/f/LkyVVVVfv27YtGo7NmzaqtrQWA77777oYbbrj++uvtdvsrr7zyyCOP+P3+5NcePHhwxowZWq123759dXV1CxYsmDdv3r59+y6w2bfeeivGGGMcj8c3btzY0NCwaNEiQRAu8LIE0RUX2GvgbB2nh3oNAGCMV6xY8dBDD9XV1T388MP33XffP/7xjwu/LDEwnbUjpKoXbNu2bcyYMf/zP/8zaNAgpVJZWFj4/PPPJ58wEDsCJnrR+vXrAcButz/xxBNDhgzBGDc3NzMM89lnn3366acAsHnzZvnMoqKicePGJV7ocrk0Gs3y5csxxpdeeunMmTMTTx07dgwhdPXVVyceufLKK0eNGsXzfOKR2bNnX3vttRjjJ554QqvVnrWd8neLU6dOdXLO0aNHAWD37t1d+pcTxPnqll6Dz9ZxOuk1uGsd54y9ZtWqVQDw0UcfJR75r//6r5ycnOQbEURXdLEjpKoXjBw5kqbp4cOH79692+v1PvvsswCwZs0a+dmB2RHICFZqLF26tKqqaseOHf/7v/9rMpnmz5+f/GwsFtu9e3fyyg6r1Tpr1qxvv/0WY7xz586rrroq8dTw4cOHDx+eOOQ4bvPmzYsWLWIYJvHgpZdeum3btgtsczwef+qppy655BK9Xo8QGjlyJABUVVVd4GUJoovOu9cAQOcdp+d6DQAghJLve/XVV9fX15MlusR566QjpLAXSJIkiuL69eunTp1qMpl+85vfLFq06KmnnpIkST5hAHYEEmClxpAhQ4qKit58880333zz//2//5f8hgYAn88nSVJGRkbyg5mZmW632+PxxOPxNk8lH3q9Xp7n16xZg5I88cQTHo/nAtu8evXqdevWrVmzxuFwSJLkcDgAgOf5C7wsQXTRefcaAOi84/RcrwEAo9GoUCja3LTNSnuC6LpOOkIKe4HVagWAWbNmJR6ZNWuWx+OpqamRDwdgRyABVsrcdNNNGzZs2LFjx0033dTmKZPJRFFUc3Nz8oNOp9NisVgsFqVS2f6pxM9Go5Gm6ccee6zNWGXia8R5e+edd5YvX75w4UKj0YgQSkzqE0SvOb9eAwCdd5ye6zUA4Pf7OY5rc9Pc3NwLvzIxYHXUEVLYCy655JLOTxiAHYEEWCmzZMmScDhcUFAwc+bMNk+pVKpp06b95z//STzi8XhKS0vnzJmDEJo+ffrnn3+eeOr48eMnTpxIfu2cOXM+/vhjURS7sbUY40gkolQqE4+8/fbb3Xh9guiK8+s1ANB5x+mhXiPDGCffd+PGjdnZ2YMGDer2GxEDR0cdIYW94Mc//jEAbN++PfHI9u3b09PTCwoK5MMB2BFIgJUy6enpPM+fOnUKIdT+2ccff7y8vPzee+9tamqqqqq68cYbWZa97777AODRRx8tLS39/e9/73K5KisrV61aNWnSpOTXrlu37tixYzfddFNlZWU0Gj169Oizzz4r5xieN4TQggULXnrppQMHDvh8vueff76+vv5CLkgQ5+G8ew2creP0RK+RZWRkvP7666WlpX6//5VXXnnttdd+//vft5nfJIhz0klHSFUvWLhw4dy5c1esWLFv3z6/379+/frPPvvsscceo6iWMGMgdoQeSZ0nOpBYBtL+qTbroTDGmzZtmjlzpkql0uv1ixYtOnLkSOKpd999d+TIkUqlcsqUKXv37r388suTVxFijCsrK5csWZKenq5SqQoLC1etWlVXV4fPtgzkjAlViSs7nc4lS5aYTCaLxXLHHXfIOVhvvPHGhf1KCOIsuqvX4LN1nI56De6043Tea1atWpWZmWm326+88kq1Wp2Xl7d27doL+30QA1TXO0Lv9wKZ3+9fvnx5ZmamQqEYM2bMa6+9lnhqYHYEhDHuwfCNIAiCIAhi4CFThARBEARBEN2MBFgEQRAEQRDdjARYBEEQBEEQ3YwEWARBEARBEN2MBFgEQRBE/1ZbW/vggw+OHTtWq9UOGzbs7rvvlmuXE0QKkVWEBEEQRP82Z84cp9P5zDPPFBcXV1RU3HLLLTzPl5WVaTSaVDeNGLjICBZBEATRv82bN2/Pnj3z58/XarVTp0598cUXjx079t5776W6XcSARgKss2tqaorH46luRWcCgYDP50t1KzojCAKp/D4QRCIRl8uV6lZ0pu93Fo/HEwqFUt2Kfuahhx7SarWJwyFDhgBAYpvhHiIIQr/bq5jjuMbGxlS34tzEYrHk/Xb7ERJgEQRBEBcVec87OcwiiFS5qLcBIgiCIAYYl8v1yCOP2Gy26667Tn7E7Xa3GVhNS0szGAwXeCNBEARBOONeSX0W3yrVDTkHfafNLMue0/kkwCIIgiAuEjzPL1mypLm5edOmTWq1Wn7w7bffljfyS1ixYkUi/DpvgiC43e5z/dBNLZ7nvV4vTdOpbsg5iMfjwWDwjPu797KcnJxzOp8EWET/9NvfwsaNcPnl8Ne/propBNGPSRgwYLr100vEGAGiUv9Zdj4wxj//+c83b968YcOGOXPmJB6/66677rrrrm6/nSAINE2f64duanEcp1Qqs7KyUt2QcxCLxbRabUZGRhfP5yXMJr2D2xz2JpKDRfRPdjtUVEBtbarbQRD9GAbY2+ArtXsECQOAKOHtDs/uBq/UP6v3rFy5Uh6sWrp0aarbQqRGhBe/qW7+3t2yTCTICV+eclb5IilpDAmwCIIgBihBwiFecEW5HXUeTpS213maI1yEF8V+WB/xqaeeevbZZ5988snly5enui1EysQEKS5KFa7gUXcoEBe21LpjghTihJQ0hkwREv3KffdBUxMAwK5dAAD798PPfw4AYLHAM8+ksmEE0Q+xFCrJs2xzeJoj3OdVTkHCJhVbkmdJ1ZTKefvnP//54IMPrl69+qGHHkp1W4hUsqjZ4jzLNofniCt41BMSJDzMrB2XcaELGs4PGcEi+pUPPoDXX4fXX4fqagAAu73l8N//TnHDCKJ/YmlqRq6FRkiQMIXQjFwzS/e/z4V169YBwNq1a1GSnsi7IlLLEYxud3gSI6wnvOHd9d42460WtWJCphEABAlb1YpURVdARrCIfubSS2H4cACAgwehoQEyMmDSJACA9PTUtosg+ikR470NPhFjhEDCeF+Df0aeme4DK7bOyeHDh1PdhItfTJBUDNXRYS/AGI66w/44v7POW5RrrvJFyp0BmkJBTjAofwhmApxQ3hwAAITAHeWOeUIjLLrebGcCCbCIfuWVV1p++OlP4d13YcYM+OijlDaIIPoxEeMdDq8zEjcq2aJc8+56rzMS31HnnZHb/2Isokcd84SOukPFNqtZxQLAQWfAEYyW2KwGRe9FEQhBic2y1e5pCse/qXYFOYGm0MxcS3J0FeSEbXZ3XJCGmbW5elWpw3O4OYgxjLSmIMbqf0PBBEEQRPfAAABGJVtis2hZuthmMatYScL9MMed6HG8hLfZ3Z4of9AZOOkNA255//QmBU2V2Cxqhg5yAgAU5ZjTNYr2p8l5V1a1ojjPwlAoVW9nMoJF9E8GA6SlgdGY6nYQRD9GU2hGrlnCWM67YimqOM9CIUT3tyR3oqeNsOhEDJWu4Ba7S8KgZKheHr5KqAvGooIo/3zCG07TKJJHW/UK5rKCNBXTUknVolbMG5yuZlJTWJWMYPU/QlKNGoyhPy6o7gZ//zs0N8Nrr6W6HQTRv9EUSs5qZ2mKRFfEGY2y6ixqhfz5MzbdkJLo6qQ3XNbkpyk0PcdsUDJN4fiueq90+oeg6vRwKlXRFZAAq9850OTfYndzogRykcBGX6ndIw7IEOti9dVXX6HTpaWldXK+0+m8+eabLRaLTqebP3/+kSNHeq2pBEEMHIecAU+UYygEAAebAr5Yb28OiDE4gjE57ypXryqxWQ1KxhPlI7z4wzm93KZOkQCrPxExDnKCL8aXOjycKO2p99oD0aggCpLUQ3esC8aOeUKJw1O+SLU/NSVxB5ry8nLcqs0+tcl4nr/yyitPnDixf//+mpoaq9U6Z86choaG3mwqQRAXvUPOwAlvWMlQcwrSRll1vCRtc3j88V6NsRCCWXmW2TarnHelpKkSm3W2zaprHUurcAX31PsS41kHmvyHnIHebGEbJAerP6ERmpFr2VH3Q1VALUvPzrfykbDU3TFWmBervJGaQIQTJYwhW6c64go2hGI0QplaZQoHXYlkb7311sGDBysqKgYNGgQAf//733Nzc9esWfP000+numkEQVw89ApGyVAleVaDgjGk6QGgJhBlqN4eo2EoJC9jlClpStk6wc2JUm0gGuFFCsHkLFNZk7/aH1Ez9EirTpmi0m5kBKufYSg0I9eioClBwghgRq6lh2KdvQ2+496QRcUyFDriCn5T42oIxRBCRbnmHrojBtjX6GsIxeRDEeOd9V5PtLdHofuXTz75ZOjQoaNGjZIPdTrdZZdd9vHHH6e2VQRBXGQGmzRXDs5IFEQYlaafmm064Q0npuTcUe5wc7Bb7vXNN988/vjjGzZs4Pku//0PhRSu5ku1SMPStYHoplNOOboqsVlSFV0BCbD6HQxQ1uTnRAm1RiR8z+zLOiXbpGHpxnDcpFQAgIQxIJiZa87UKnvidgDgDMdr/NHd9b6mcFwuz1MfjB12pXKAN4Xmzp3Lsmx2dvYtt9xSV1fX0WlHjhwZMWJE8iMjR448depUNBrt+TYSBNH9eAkf94QTh3FBOukNd3J+r2mzgdIhZ+CkN7y3wYcB3FGu1OE55gk5I/ELvMuf/vSnyy+//Pe///3NN998zTXXdLVkyCOPQFaWevGi2TYrQ6EwL1IIFdssulRk4ieQKcL+BAPsa/DZA1ENS8/Msxxs8jdHuFKHe5xR0e2RspalS2zWrXa3K9raYTB4Y3zPBViZWuWkLGNZo39nndegZLwx3qhki3LMXb8CBkAdH/YXSqXy4YcfXrZsWU5Ozs6dO++8886ioqL9+/enn6lavcfjmThxYvIjJpMJY+zz+dRqNQDE4/FYLJZ8gkqlYlkWeobUqoeuf+EuyhaWNwfTNIpsrRIAMIYyZ8CmV52xPtD5oXp9Jmgg29fgqw/FQrwwMdMYF6StdneAE1iayjeoU92000zPMW+1u+2BKCdK7ignSPiSdH2GRilh3BzhEp8UgoS9Mb6L70aM8R//+MfE4RdffLF3796xY8d2vVXHPCF5ob2EcaUrNDXHlMJPARJg9ScYY0HCGpaebbNqWHpmrmV7nYcTcQ+VBZQw5lv/yptVTJATK1xBCkHPbTswyKiRMBxo8ntjvF7BlNgsinMZ3T3Q5McYJmYZEYCEYXe9V6ugx6anbCOq81NSUlJSUiL/PG/evA8++GD8+PHPPvvsH/7wh/O42jPPPPPUU08lP/LAAw8sW7asGxp6JtFoNBqNchzXQ9e/cKFQSJKkSKTvrtXw+Xwsy2q12i6eH5ZQVZSq8oWHqbCJwifjlEdEzaHIGLXUXR8teXl53XQl4uzGpOu9Mf6ULyJh7I3yAU7I0irz9KpUt6stDUvPzrduqXU3heMAcEmafqRFBwB7G/yOYHRipnGwSSNIeEedxx3lZuRaNCy9t8E3Ndskjyp5otwBZ2BmrrlNVQWa7uywcxFerPJFVAw1Oct0oMnvCEbpRpiUlbIYiwRY/QmF0LQcMye27ABFU2hGrkXCOBYOSd09URiIC1vsbl7ECJCCRt6YkK1VNke5792hfIOmh7agkjBubM3BCvOiJ8pn6bo6YBYTxMZQXC5ANyHTsKve1xCK6RVMoRWz/bmuz9ixY202265du874rNls9vv9yY/4fD6EkMlkkg/vv//++++/v8db2SoSiUQikc7rSqRWIBCQJCnx++mDNBqNQqHQ6c7ha4wmEN3b4DsZRxaVwiNy8vCzhiUrUfolnYKRN4Sp8UcBIFOrnJ5rpvrkzkVRXoyLLV/C/XEeY0AICozqhlCsrMkvYtwQijVHOJOKtajY7z0hb4zfaveU2CycKG1zeAQJ14fiQ0yaxAURQo899tg999wjjxpcd911kyZNajMGf5rrrwf5b6PPBwCqw+XXTBvFACCNurjWsdXu9kR5XpTO6Yt6NyIBVj9DIUgObhgKAaCO333n70CTnxOlfIPaZlDrFMyWWndDOH7pWy9r3c2qq6+Cq67q9juKGO+s8zaF40Ylm29QlzcHdtV7Z+SaM7o2KaliaPmvUrU/0hSORwVRr2BKbNZ+HV3JEEIdDVJecsklhw4dSn7k6NGjgwcPlucHiQHCZlBjgL0NPleUUzMkuur3WJpiaRQVAAA0LN31fSEbGhpWrFixdevWYcOGrV27dsaMGT3XSE+Uk4OkERZdXTDqCMYo5JucbcrUKotyzTvrvHKJBJOSLc6zsDQ1Jt3Ai7jaH9li9wiSJEh4dJo+ObqS3X333dOnTy8tLR02bNjVV199lkY0NYHDkTiieI7iOQCAYED+mkEjlKroCkiARXRkWo65NhAdYWmZp5idb20Kx63vvQMVFWA29kSA5Yny8qaz8swgQ6GyJv8Jb7iLARYA6BRMsc3ydXVzVBAZChXbLL282XtPOHz4cG1t7dKlS8/47OLFiz/88MPKykp5IWEoFPrmm29++ctf9m4biRTDAM5wS65kTBA9MU7Dkgi7P9m4ceNzzz0Xi8V+/vOf/3TZzVvt7kBcyNAqA3HhlC+CACZkdmlbsGXLln3zzTcA4HQ6Fy5cWFVVZTafQxrrOSlvDsp5VyMtuiEmjZyPNcioSdMo0tQKOY8WAGwGtRziyP8KXpLqgjEAGGbWFnawAfP06dOnT5/epUbceScsXgwA8OmnsHUr2GwQDoPHAwsWAIA21V8zSIA10AmCcOTIEYvFYrPZkh9XMVQiugIALUu3/6rRvdI1ipm5FrOKlXvjYJNGxVAZmnPIqccAla6QPFkqSLjSFZLzsfqXW2+9dc6cOSUlJVarddeuXXfeeWdOTs4999wjP/v5559fddVVb7zxhpxHddNNNz399NO//OUv3377bb1ev2LFCpZlV69endJ/AdGr5LEreyCqZekhJu3h5sDeBj+NULauz2XtEGdUWlp6zTXXyMsaNm/erB45kTGnZ2mVRbnmMC9utXuqfBGrWmE7W5K7JEnffvtt4tDn85WVlV122WU91OyiXHNDKDbIqAEADUuX2KyeGJemUYgS3l7n8cZ4DUNHBbG8OcDSSMPSkoQPOANcazZLXTA21Ky90BjoxhsBAAQBduwAANBoQK7skJUFmzYBAGRmwvjxF3SLC9Dvv98TF+Lo0aMjRoyYMGFCQUHB7bfffuZ5KFGEDRta/pPTfQ4fbjncvPmstwhxQieHbWRqlcnDudk6Vde3RcMAe+p9jmBUp2Bm26xqhq72R8oa/Wd/ZR/zwAMPbNu27bLLLrNYLL/4xS9mz569e/fujIyMM57MsuyXX345dOjQiRMn5ufnu1yub7/9Njc3t5fbTPQEXsK1gR/KbcSFlq/+bfhifF0wJk+IDLdoJ2ebMMbfu0N9as8QohPvv/9+8qLRD59fN8SkKco1UwjJa32GW7R5XVhCSFFU8ppilUp1ySWX9EiLAQCAoZAcXcmUDJWnVwPAngZfc4QDAMBYXspd1ugvtXu213kjvCiI0ug0/SCjJiqIW2rd4aRdbs5fIAAffggAcPQoNDYCAPzznzB/PsyfD4891g3XP19kBGtAW7ly5alTpwAAY/zSSy9de+21V7Wf++M4uPnm0x756CP46CMAgAULYO7cTq7fEIrtqveNStMNNagAoMYf2d/kn5hpTO6W3YUTJV9cXntoVTGUnI/lj/OChJl+lYY1dOjQF198saNnFyxY0CYOzszM3LBhQ8+3i+htu+q9znA8JogjLDp5uX6QE2bmWdqUSjGr2Jm5Zp2CkfOu8g1qlkJWtaI/vekHtqFDhyYf5qSnJU8I6hVM15dCv/XWW8uWLduzZ09OTs4LL7yQmZl51pf4/f7vvvvOYDCUlJR0fcleiBO2OTzj0g05ehUA+GL8jjrv5GxjhkY5wqLlREnE2BvjVQwdE04LoUan6QutOvlPWLU/4o5y2ot3OpsEWANadXV18mFNTc0ZTkIIEr1OFFsekevinK03RngRAz7SHBRFieehqtEP0FMph0qamm2zACA570qnYGbnWxU06l/RFUEkjLLqPFHucHNQkHB9MCYv1z9jPaE2eYpkcrB/ue2227755psPPvgAAGbNmvXwww+f96VGjBixe/fueDyuVHYpueLQoUNXXHFFc3MzAJSUlGzatEml6tKbxxcXory4u8E7Dcxqlt5md/MS9kT5DI3SolbMzrcKEt5e53FFfqjYMsqqS9O0vIERwMQsY75BndYt1drMZnC7W36eNAlqauDhh+HeewEAFO2u/957gDEUFcHpWTE9gUwRDmjJqdMGg+HMSzZUKhCElv9GjwYAeOghEAR/OAqffZY4q/2unxig2h+VF+5+7wmfjCEMAIBVPbaPoYqhk7PatSzNkgKJRL9lVStm5VkYCn3vDgU4oS8v1ycuhEKheP/992tra48fP75t27azpqVLkuRyuUSxw8m1LkZXAPDII4/I0RUAbN269a233uriC/P0qinZJoxhd4N3q93NS7jQokuOlhgKtamMWuWLJO9agwC6J7oCAITAYmn5T/6br1a3HLYvd3LDDXDDDbB1a/fculPk42dAe+CBB9avXz9//vwbb7zxu+++s3U5oj/li3xd7TrRuoFDpSv4TbXLHjhtexYEUFxbOeWp/75k7R8Sj0zJNllUPVVGnCAuMjoFk9j606pWdH25PtHv2Gy2YcOGnfW07du3Dx06ND09PTc3d+PGjV25snh6lcTkI6/Xm/xUm8POeWM8BpAwCBLO06uCvLC11l3fWsjQEYwmp8BqGDouSvLqyK7for/rQwFWbW3tgw8+OHbsWK1WO2zYsLvvvtudGPQb8OQcqYULF95444179+49jyt89tlnhYWFSqXyqquuqq+vlx+kKOquu+76/PPP33rrrQkTJnT9ajoFw1DokDNwwhuudAUr3SGGQu13fVIeOZz/2suD332j5V8BEBW6I6WRIAYA+QMpyAlpagVDoQpXMHmLOmIAEkVxyZIlcmpHU1PTT3/602AwCAAul+vGG2/MysqaPn168kJCAIiL0uZaV6U7JB9GBelACFf5WnYyuPPOO1Fr1G61WpcsWdL1xugUdGIzsrpgrC4Y07C0WcUCgDMS31Pvk58dn2m0qNiIIGpYmpdwpFuy2jvx+uvwxRctqwsTOA5OnWr5T9bc3HLo8fRcWzosYNj75syZ43Q6n3nmmeLi4oqKiltuuYXn+bKyMo2mZ6sDnFVTU5PJZOr6oGtPWLduXWLtvVarPXTo0JAhQxLPti1O3dQE8TgYDND6iMPhGD58eKIe7rx58zbJS1jP1dGjEI1CVhZkZcm7e8q7PrEUmmWzth+a8j7zV/PK3/AGw6e7vwcAhABjGJdhGGbu6jYgRP9CKrlfOI/HI1dy32p3N0c4ebm+N8bL3W1WuyR3IlUEQWhqaurNRbsul6vNnqSHDh0aO3bs4sWLP/30U/kRrVZ78uTJRHq7L8bLU3ij0/T5BvV3ta6oINkM6qnZLV3gq6+++ve//20ymW6//fbBgwd3sSW+OL/N7uFEiUJIwhha06rkBUwixrvqvQUGjV7BGJQML+GddZ5Cq15JUwblOWd+x2KxQCDQ0UrqLtm/HyZPPvNT994LTz99/lfuVB8awZo3b96ePXvmz5+v1WqnTp364osvHjt27L333kt1u/qE5KnxcDj8ySefdHb2okVQUABPPJF4oKysLHm3ge3bt59nO7HjVdUAACAASURBVEaOhAkTICsLAKxqhbwoFwAKjJrToqsdO+D11+H115k9uwGA4viZX3869JN/2z78tzIWNijJFCFBnN24DEO+QS0v15fzsYaYNF2vu0tcBMrKytavX//ZZ5/JdRzS0tKmTp2aeHbw4MGFhYUAIBcXlYXD4d27dycOTSq22GZlKarCFfy62hUVpHQFmpL9wxeMK6644qWXXvrzn//c9egKAGKCJEi40PpD3hVGkBiuoRGamWvJ1avkcIqlUInNmq5RyIdtpizF7t7nre/oQ6sIH3rooeRDeYTmzOvaBp6CgoL9+/cnH57TyydNmqRSqRIx1qxZs865BT/+MWAMq1ZBSQkAeGN8Uzhe7Y+wFJIATnjDWpYemhiXevllePVVANADAAAdi2Yt/3UWAAB8OXmaYuS5NZ4gBiajkk3+ILSqFVa1AgCgshKKigAAdu6EUaNS1Dqix7300kt33HGHPMv04x//WF5m+OGHH65atWrv3r2jR49et24dy7IAMHny5C1btsivYll23Lhxydcxq9jJ2caddV5ekgwKZphaPGsqHwaocAUHGzVy7Q8JwxFXYIRFl8hSz9IqLx+UVukKOcNxDUvHBUns2nQYL+FSu9uqUci1J+R5cJtePbKDqu6dc0e5497w1GyTnJ7YGI7bA9EpWaa2yYqFhbBzZ8vPM2YAxvD443DllQAA2dnncd8u6kMjWG18/vnn0BpmEWvWrEnkP/7iF7+49tprz+nlubm577//fmFhoVqtXrhw4b/+9a9zbsFHH8HHH8u7Ph33hjfXuCpcQZZCxTZrtlYJAAedgWp/y7w+MAywLLDsD3UcWBYzDLDspGyTiSS5E8SFEEUIBCAQgI7XkREXgYcffjgRtHz44Yd79uwBgNzc3HfeeefEiROffPLJ8OHD5Wdfe+21mTNnAkBWVtaGDRvafAOPCmK5MwAACCDACY447G/0+2It675FCe9p8AVPLwFd648edYe22N0RXpQw7Kz3HPeE97cmrXOiBAAnvGFHMKpl6UvzrTPyzDRCB5r8jjPVwm3TmBAvHveEy5sDMUHaUusOxAVPjO8oOGszvNXm8HBzsD4Y21HnFTGuD8V21nkcgagrykEbGg1Mn97yn2zo0JbD/PzOG3wh+lAOVjKXyzV+/Hiapo8ePSpvW7tz586diQgUAACKiop6tExtgtPpNBqNqc3BAgBRFCsqKoxGY367N0QwGJQkKXfSJHmIFvl8IAigVmOtFgDE6dOjb799xmtyElYklYlqc5hMbzAAQOzVV/nrr/dyYpk7KmKwaRUU4Jowz1CIBjwpTaM7fe8/9uWXVStXYqPRd+pUc3Nzdk9+V+givV6f6iZczAZUDtb37pBeweTqW+oGlTcHMjXKC5/CS+RgdXjG4cMwdiwAQHk5jBlzgbcjzluP5mBhjM1ms9//w0K80tJSOYrqyBnLXyUKptcc2vfdB+8sffBJRqEEAJamivMsBgWzvc4jp/rNzLP8cHeA/Y2+Gn9Uw9IshfxxQcPQcwqsKoY+5gkd94RLbFYJ4/1N/qIcszzK5YpyR5qDM3LNZ6106I/z2+yeuCixNMWLUrZONT3HdMb6IxKGXfVeFuF8RsjIyJA34UnTKEe1DndxorTN7vHFeZOKDcR5jGFSlqnA2GnlUooCjOHNN6GDDV67UR+aIkzgeX7JkiXNzc2bNm2SoysA8Pl8bapiFhYWCkJvLPgURVEQhK6XuO058oa+7f/VgiBIkoRay5m0iEZRNAoA4PWe8RfVEBOPB/lRBtasoBkE1WGhNiKMMyoNCkQBcBL22Rvz//1m8kuoTz5hTpxIByiaOHnnuBn2MAcACgqNN7EamqJAEgQp+Xy6df8HQRDkX+OF/QIIoq8IcEKlOwgAU8Gcp1cddAZOesN1wdiVg9N7qlTVpk3w5z8DAIRb1xLeeitotQAAv/tdy3wHcbFACD3wwAP333+/fDh37twieV64Y2ccBeAlzIni7i83rrnn15IoVlRWPvHavzNNBk+M32b3GJSMO8qZTp+MBgAEMDnLRCF0qnW9ISdJYV6s9kcrXEGGQoIkWdSKywp++CqVplZcmm/tyj/NqGSn5pi22T28KGlZuqPoCgCiguiJcXFBElSURcLbHR5XlBMkPMKilecEFTRVbLN8V+uWB+QmZhrPEl0BwHPPAcaQlMrWc/rcCBbG+KabbnrnnXc2bNiwtOcDzK7oC6sIO9fypfyNN1qSDNetg9pamDsXfvQjAIC8PLjuuvavOuYJHW4OIgCtgsnVqY56QgDAUsisVkzKMm53eKj9+y/7yYIz33Llyr2rH5E3Shtm1o7LaN3JYfdueOQRAICPPwaVCl54Ae68E0wmobnZ6XTm5OR0+7+d6FMG1AiWIxjb0+AFgAyNsikcVzJUic1qaFes5Fx1OIL1r3/BLbec+TX//Cf84hcXeF/iXPXCKsIvv/xyy5YtQ4YMWbp0aVc+g/xxXssyie0rvDHeqGQ+/s//Xf+jxVLrhPKQwtEnK49UuILfu0MAYFAys23W9sNOiU3EAYChkCBhecEgQ6HiPItFff41QhMFsRAABhhh0Y1J73BiIcgJW2pdcRGrGComSGYVW2yzJFeQbgrHd9R55ZWMGVrljFxz3ykX1+dGsFauXPn2228/99xzfSS66k9WrGj54Y03oLYWJk6Eu+/u5PQRFp2EocIVDHHCUU+IoZAoYV7Cwbiw3eEJxIXBJj0kNsk6eRIAIDNTLozbpNbXBqIKmpIwPuENq1l6uJzh7nS27GEuD1YtWQKzZgHT595mBHHh8vQqBOZd9d6mcFxBd0901ZkRI+DWWwEAvF54/30AgJ/8BOSq3yNG9OB9idSZN2/evHnzuniyP85vqfUYlIy8AYAzEt/h8KZrFMMHF+CkzaRNGpWEcSIHK8qLYV5sE2DJJdrrgjGdgjEpWUcwSiMkYgwAY9IN3RJdZWqVo9P0pQ7PMU8IA+5ov0W9gpmaoS9tCMQESUFTxXmnRVfydrcY4/EZhmp/1BmO76zzFvWZGKtvffI99dRTzz777JNPPrl8+fJUt2VAqNq15XBzZEzRLAAQRAkQohCKCmJUgEytctzcGXDiBMZ448aNVy9aBADw+ONw3XWVElvpjShoqsRm4SW83eEpdwbY03dWbyFvVgCt8RZBXFy8rZ9SvCj5Y3zPBlgzZ4KcgnP4cEuA9eijJAeLSNCyjDzlV+rwDDdr9zT4RIwztcqheZesXbv2oYceisViQ4YMWfv009vrvM5w3KRk07WK455wqcNTbLOYkgro1AaicnQ122ZRMnSoWvDFeXnA6XBzwKRkLerzWauEAThR4kUs510BQLHNss3u4cQOZ9JECVd6I/LTnChVuEPjM34IxU56IxhjOe/KZlBvtXuaI5wvxlsvIATsRn0owPrnP//54IMPrl69uk29BuKcXX01jBoFU6Z0fpYvGneo08cU5QkcxygUgBDGWKugg5wIAFqWlr8ELFu27K233mp5+//61/DrX9t27XGkF0zLMRmVLADMyrPUfPhZbvleoBCcONFy9ccfB5YFALj1VjiX8ioE0V/IOxkoGWq4WXvEFdzb6KMQSuS8E0QvYyg0K8+yo87THOHcUQ4AxmcY5Oo5K1euvO2229xud1ZWVlWz7/tw3KRki20WBU3RCH3vDu2q9+bq1GPS9fKfemcknq1TTsw0qhj6qDvki/MUgpm5Zk9MqHAFSx3uYpvVfI7rwTHAvgZfXJRK8i0ahgaAXfU+23PPzmeA+dFiyDpDIVA5q90dE3QMmpJj2VnvPekNI4BEUkpRrtkV5bK0SgCQv/P740Ifia6gTwVY69atA4C1a9euXbs28eDy5cufe+651DWqf3r00a6ctbfBl5aThyXs97itWdkAgBAKcqJVowhxQpUvQiGk8NS33/5Tp2CuGJyeGIG1qhWWY4fQn5467aQ1a1p+mDfvPAIsUcJ00nrGNocEkXIBTjjliyTyrjQsvafBV+EKZutUPf5WVatB3thKfbZ8XqLPE0Xxb3/72xdffJGdnX3fffeNuLDZXoZCw8za5ggHAFqWLkiaVdDr9Xq9nuM4CwtTs02ZWqU8LTg6Tc9Q6Kg7fMwT8sQ4GiGagvpgXKdgvneHRYwtapalqFl5FouazdAChdBxT+g85uAEUfLF+UBcOOQMTMs2727wNobiU9c/zYRDYMs9Y6V1TpIivGhU0KM02KJWFNusW+3uACdIGMt58QyFspLW7SpoKr27NpDuDn0owDp8+HCqmzCwDLXo9jX6KIqyZmWLvECzrW8GDLNyLaV1nhPecLbUsnbSBDACIFEeuE3fQpmZLVMVwSDItWFHj27Z1byT1eYd4CVpm92T1lqJLiZIW+3ufMN5VqIjiJ5gUDAz8swqhpanBfP0agohk5LtjS8CQ4dCWVnP34boDY888shTT7V8O/30008rKiqs1s7W4kWj0e+++06pVJaUlDDt0lubI9zuBh8AaFk6zIulDo+cj9XmNJvhtNB8hEWXoVFuc3hckZYKUjoFY1GxVb6wkqFGp+lyh6gSSVojLNpBRvVZazG0x9JUic261e5uDMU/r3JyoqRXMO3blkzN0LPzrQIXj4ZCAGBQMJfarCqWTl51iNt9HvUdfbfQKNHTBpu0JenqaCAAADTLbHrn9a2ffaAUYnoFY1SxEzONZhVbNHr4n4qKqgEqFMrPNa2F2hcuhEGDYNAgePfdlkfuuAPKy6G8HBLDjbt2tTxytpnK9qK8FObF457w4eZgVBC32N1BTvDF+W74NxNE10R4UV4/JQvEhfpQ2wqKGRplctJVjk4lFwQiiK7bsGFD4men0/nFF190cnJNTc2oUaOuuuqqyy67bMSIEW12OvHF+O11HlHC4zMMlw9Kt6oV7ii3o65LmxmbVOysPEsiUhExrg1E5QFaNUO3CacShxzHrV69Oj8/f9y4ce3nOtpT0tSsPMvwN16Z8l9Lx//xv2cf3NGSnlteDp99Bp99BuXlbV6iZujkrPaYKH1b40rsGF0XjH1b45ILn/ZBJMAa0JRavdHcsl59yqWXzx4xaMHoQZOyjABQ4Qp6Y/whZ+C3d9xRAJDDxS2R1uo7DQ1QUwM1NRAM9kSrDEqmxGZR0NQxT+irU64QJ+ToVVOzzT1xL4JoD2ModXj2NPjkIkCBuLDV7t5d7z1rlM9LWN6hWT6Mi9J3te5EIjxBtGez2ZIP29eRTvaHP/xBDqoQRVkLho0ZM+bEiRMAIGHcFI4blEyGRinnXcn5WOkaRUH7tUdnggFOeMOJVPMoLyIAvdt+w+JF48aN+93vfpe8m23C448/vm7dOrvdXl5efvPNN+/YsaPzu0gYH2gKZH39edaWb4a+/rLyx9eCXKzxuefgmmvgmmtg/frOr1Djj8j9MSqIdcHYngavL8a725du7xv60BQh0cu8MX6b3cNLUqFV548LkJ2rH1QgSFhObp+eY95qd5/0hpFtOLvqQQlDTixofn49AMDdd7fs3zRtWg+1zahkp2abSh0eXpJ0CmZatpmkYBG9BiEYl2HYWecta/JHBfGULxIXpaFmrels+5Q3hmLNEc4b88zMtegUzFa7O8gJJ73hNlUcu0siE+WMh0Tvq6+vf/LJJ0+cOFFUVHT//fdrNGcPbtavX3/11Vc3NjYihFasWFFcXNzJyW63W/7hV4/+6coblr31zJ/Wrl372/vuOyUqAxI1NdtUlGtOvAMYCpXYulT5E2PY0+B1BGM6BaNXMA2hGABggK/Lj3+zeTPPceXl5aFQ6G9/+1ubF3799deJnyVJ2rx584wZMzq6i4Rhd72vIRQbeQHv0kmZJlECRzD6bY07LooYw+RsU7auj64sISNYA1eNP8JL0qg0/eg0/fQcU7ZOFYgL37uDcn0UvYIpsVlZCp3IGVR5213sQw+YV9zZ8spbboH774f774fTtxQFAFi4EGIxiMXOI/UqWUyQDrVunhVqLZlNEL0mU6uUq+l87w7FRWmwSZO8OLwjNoN6XIZBkPD2Oo88tZ2tU8lDwt1OwlgOAeVDXsJbat2V7lBP3IvoCp7n586d+8ILL3z55ZdPPPHELUlVYeX9AZVK5Y9+9COXy5X8qkmTJp06dWrHjh2nTp36y1/+0vktfvWrXyGEJpbMLbn6x/FIZOm99wuW7A8OHA9IVNDj0sLZd3E+o6gguqK8TsGkqRUNoZiCohiKAiyOL55zz7rnaYYFgE8//RQAMEA8aT5u8pSpOsMPb+9JkyZ1chf8ztvmt14v/OhdczgAANKUKd8//idJrp76179CMAjB4FlHsBCCqTmmdI0iKogShjEZhnxD313qQUawBq7xmcZMrVKO/SmEpueYTvkjh5oCtYFYcZ7FrGKrfGG+dWvNENe1bWUpCi645H1ckLbZ3UFOyNGrRlp0pQ7PUXcIYzymg0p0BNETNCzNUEgUMQCcdewqYZhZy0u40hUMcUKaWtHRNiDHjh07ceLEhAkTznt7g6ggeeN8YzhOITQ6TbfN7vHGeAqhkRYtGcdKib179x47dixx+N5773Ecp1Aojhw58otf/EIURQD4+OOPlUrlu4nsVQAAUKlUZ90GRzZ//vw33njjnU3fqjQageOi4fDCn/8KADDA6+v+aJ9wycMPP9zmJRjAG+USpUEljEPt/pZrWHq2zSKXbpfzrjDGFYeP7HUGBxdeojeZfS6nvFHbwSZ/YzheYrNqWRoDLP3tI5NvuGX19Qu5SOh3v/vdggUd7PwBAAD0qlUjGxoSh9TevYV797YcKJVd/07eFI67oy3T7ie94dw+nPtIRrAGLgSQPLJKITTUpC206nhR2mb37K73nvRGACBPr1IyVLU/Uskh+MlP4Cc/gfPdZiR0+obtbQ4TeEniJZyjV03LNptVrJyPxUt9a08n4uIW5IStdndclHJ0KhqhA03+an+kKy+Mi1JdsCU73hvnE58EyR599NHCwsKrr7568ODBb3ewEXtCYygeS9risy4Y4yUJALQsPdtmVTHUSW94U1WzN8anqRUz8ywkukqVvLy85C1rs7OzWZYFgD179ojiD0HNzp07L+QuN910018ffbDpwE5GoVS1TkHu/vLzb95/p6qqqv35ZY3+72rd8ooNCeN9ztDhEE6kCSbIk4PFNstsm9WgYIxKdsbkCdyxA3+4banP5Rw7duwLL7yAAQQJR3hxq90d4oS9Db7GqJCbl3ey6qTP52sf28UFaX+jX2j90y2pNZJaDRoNyL8lmgaNBs7x7doYju+s82KMJ2eZbAZ1hBe3OTxRoWvf/3sdCbCI04xK049J1/OS5AjGAGCQST0txzzbZlUy1PdKg2fD2/DeezBokHxyZyHPhx/CrbcmV+RyC/BldfMJb0um/Elv+MtTzbWtC7W8Mf5461M6BTMmQ5+pUcp5V0Yle1lB2sTMHplnIYj2MIYddd6YIA01a4tyzUW5Zgqhska/vzXJvaNoP7ENSLZONSZdL0p4R52nTQZuTU3NE088IW8Cy3HcXXfdJUkdroFyRbkd9Z6tdrccY53yRXbVe7c7vPKzegUzI9ci31fFUDPPtCCf6DU2m23dunVy6QSj0fjqq68ihACgqKgouZ5C51lWTqfzkUceWbZs2RVXXKHT6Uwm00MPPdRmy+DBgwffvuQ6rYKWr89z8enzFvz4V3cdPnz4pZdeanNyplaJEOxt8NX4o7vqfU0RTkmBvoMtB1iKSn5q1W9WHDty2G63Hzp0aMiQIQhgcrYp36CO8OJX1S57IKph6RKbxaTVyJFlm0DnsCtQ7Y9sd3gECYd5cdOX2z8qO3nS4YSFCwEAbrgBwmHYsgU2b4ZrrunKbxgAmsJxjPHkbFOBUT0ly5SnV0V5MdjBd/WUI1OERFuotX/SCA02ajFAc4QrzrNU+6OHnYHpuea6YGywSXPSG24Kxzvc9WnvXnjlFRgzJhFjKRDQCB1yBkQJK2jqoDNAI6RiKACQMN5V743woiThkVadMxLf3+CXMLaqFQYlAwB9dgSYuCghBFOyjPWhuLwHrSBhJU3ZjCp564Jqf6TKFynOs7QvBeSKcMG4IG8DQiGEAJU3B+yBaHJpaZfLlRxR+f3+eDyu7qBkqEWlyNaq6kMxuRTcEVeQRqiwtSAcL+GDzoD8c0yQjriCXUkUI3rOb37zm6VLl8rFFLTalro2hYWFb7/99oMPPtjY2Lhw4cJOSmdHo9Hi4uLjx48nP/jHP/6xsLDw5ptvTn6w1OFOlCpACMWjkWUrH1x//29uv/12n8/3u9/9LnFmrl5FU5addd59jT4A0LF0oUqU//B2hUajSU7VRwCTsozNEU6OpWKC1BCKDzFpAMARjO5r8E/NMeW0ToyMyzAGOdEV5bY5PDFBjPBivkE9xKw97Qadhpvtjc8w5BvUchF5hGBqttkf500d1JT3x3lexGmtpUfdUY5GqKOTewIZwbo4tZlQ6+L82quvvnrD7Xcfag4KPJfOgojxNoenrMF7oMn/yc4DdqfbFeW+PNV8oMlf6vAccgZcUS4Y7+pXBz0NM3MtDIWOuIJlTX4aoRm55gyNEgAohKZlm+Sn9jX6dzi8IsbjMgxydEUQvc+iVsjRFQDUBWMRQawLxqOCWOWL7G/0B+NC4ExfmnP1qlk2SyLvarhFOyvPMv70wdeJEydOnTpVazAMGzsBAG6++Wa1Wh0TpMCZuhKFYHqu2WZQBznhiCtIITQtx5SpVQIAL+FSh9sT5axqxdyCNHmu8FBrvEWkSnp6+pQpUxLRlez6668/duxYIBB45513zOYOK8588803baIr2a5du5IPHcGYK8IhgMlZpomZRoZVUDRddaR899dfAMCbb77Z5uUZGmViWxvK53z17y+9++67PM8DgHT6cFebw/Ywhr0Nvqggyt+rMcYHmvxVvogjGN3T4JMwTr4CSyG5/rsnysnR1eRs04UPsSZv0YMQdBQwcaK0ze4prfPI86HyFo1bHZ640HtFs8gH2EXIH+e32T2j0/SDTRoAcEe5HXXeSVnGnDOtZXVHOZ2CUdJUeXn5k2ufXvPBJi4ee+r2nw/LTv/D31+rdIcOn3L4vN68YcN3bdk8fsYsDhQUAmc4TlNoZq6l7Zv7P/8Bux0AWipNezzw4osAQEkSLF6cplEMMmrkWcJ8ozojaYsDi1ox22bd5vDU+COQtIUWQaTc1GwTBlwXjG2uccUEiUaoKNecdvp+Z5WVlUeOHBk/fvzw4cOTH8/Utl3zQVHUF5s2/aeiRqE3espKb126RN6rIC5Kc/KtunZzNwggTa2Qc2jUDGVWtdxXkKS4IMl5VwyF5F1EwryI8bmmtRB9RVpa2hkfLykpafMIhVBirOjQoUNxQ/rra56IBAPQrpKWhGF3vdcd5TQsHeZ4n8r01Z6yxx57bO7cue999n+7G/3TcswWFQsATeF4WZN/Zq6lo2+2GH6o5lCcZzniCtoDUQRwoMmPABCCaTnmnNP34uRFKRHQhHlRlDBDIXj+efjzn8HY8sUjLkpxUUrU7I3wIgbQXvCshYKmCq26g87A9jrPaKu+0h0UJHxJmk7Z5dG7C0cCrItQVJB4CR9o8iMEOgUjT4HX+mPZOpX8h1fEuNYfHWzS+GL8NodHy9IlNuv+/fsz8wowlppqaw7v3u7LzR2Vpnc47L+97SZ3U8Pv//nuxNlzAYBCSP6OMi3bdIZdn/76V9i06YfD+nq44w4AoADQwoVVvsgJb5hGCBCc8kX0CmZYUhQlJH37ISntRN+BEEzLNn/Lu+SSoRMyjW3CpieffPK///u/McYURf3lL3+56667Or+g2WSaVsgebg7kzLi8KSpUuIJyQYczToVX+yMHmvw0QmYV64pyW+1uOSdS3kWEpSg578qgYObkp6kZkuPen0R4Mfl/+thJU2699daXX34ZAFiW1Wq1Wq12xYoVS5YsSX5Vnl6Vpk5XMS0vXFwyfflv7i3fuQ0ABg0aJO/qCwDeGG9WsfsaffWhmF7BjMsw3PqrX//soSd+s+a5cCCwefPmnUe+jxkySu2eYpuFE6WddV4RY0+M6yjAQgAqhtYpmBKbRc3QU7JNCKApHI+LEgYoMGraRFcRXtxid0d40WZQhznBHeW213lm5lqYvLzEOYKEt9rdcUEqsVkNSkZ+CQDMyU/r+jxmR4aatRRCZU3+8uYAAIxO0/fyfmtkivAilKVVFuWaEUJljf5Su0eQcJqGrQ9F9zb4MIDYWj7nmCdkUDLpaoVcGHfUzEtXPvMCYPjoleexJM2dOxcA9LFAVUV5yO9vqm3ZkyERAx31hIX2YZBOB0YjGI0txRooquXQaHQK6EDrzODMXAtNoUPOgFwpGwCaI1ypwyNIWK5BXOEKHvOQij5EX1ETiPhaC7If9YRiScm8TqfzsccekzOLJUm67777IpGzrzccYdHKFbN21XuDnJCpVZ6xoENzhCtr9FMIFeWai23WHJ0qyAk76luS3NUMnZzVrj19jzaij3NFua+qmxOly+qCsU2nmh9Y85cDBw58/PHHDofD6/U6HI7khKqERHQFAAih5//6bFVV1Z49e44dOzZy5EgAOOENb65xHfeEB5s0JhWbrlFsd3holn363juqDh86cagMAJQB16g0PS9J2+xuObqakGkc1Gnl93EZhrn5VjVDAwACyNAqOVGiECCEanyRxN/zI67g3gZ/pTskzwxOyTZNz7VYVKwrwiXvQAUADIUytUp5dUhDKCYHZFa1orvGmYxKhqYQAFAIWdW9vQ80CbD6vbpgTEya9rYHohhDllY5Kk0nh1O5etWULLOWpe2B6N4G3846r7ypQoFRI//hlkuMOiQlTTOfvPhM5Y4tP/vZz9avXw8A06dPv3rRNSufeWHaFQt8rmb5FhYVm6ZReKItIdFprXn/ffD5wOeDVasAAEaPlg8Fl0urYtUMPSPPnKFVpmsUKyAhBAAAIABJREFUxXkWNUPLM4wSxvsaffIWWuMzDMV5FoZCR5qDfXZtCDGgnPJH9jf6EUIzcs02gzrECV9VuyJ8y5szGAwKwg9v1FgsFo1GO7jSaWx6deI7+vAOilelaRSDTZqiXHOmVkkhmJZjzjeox5FM9osCgxCFUKUrWOkOybu+YIwZCo0fP37x4sUZGRnndLXBgwdPmTJFLgwBACYVy1CovDngifLpGmWVL8LS1DXzLtv15cYHfnpNKOCfPXv2nDlzRll1BUY1L2ER4+EWrZyu3l5ycVGGouS9/xzB6L4GH0IwOcukoikMUNbkr/FHy5sDR90hezBSYFSPyzBMzjYdbPLvqPNMyzGPzzAMbneLsemGQqsuLko76rwRXszTq6ZkdUOqFgB4otw2h0eUcJZWKWG8o+6H3ax7B5ki7N9qA9G9DT6Lkh5nUgDAkebgUU+oOcIVGNVH3SEAQAjqg7EsrWp2vnWr3SN/e9ArmOI8q5KmAIBCKN+glvdG0CiYfzyzRvnXdYnrI4T+9Mobx7wRPhTIMOpECilpyhPjh1l0EoYgJ0R4sYup6BoKrhySnlhyaFUrEocUQjNzLe4oJ/c9i1pRnGeJCGJHa4kJojfJ38t1LJ2uUaZplM2ReEyQjjSHpuaYAGDo0KELFiz4/PPP5ZN/+tOfWq1n358kLkhy8QWDkgnEhZ113ll5lvbfsBHAhKQceQpBD+26Q/Q+eX/lbXZPpSso/1mUiyCc63WcTue7774bi8V+8pOfDBkyRH4wTa2YlWfZZncfbg4AAB+LFRroRTfdOCI/b+PGjaNHj77hhhtomnZG4o5AyyaD1b5onl5tbpczfuD04qJ7G32+GC9vAs3Q1MRMQ55ena5Rbq5xRQWxwh2M8qKcSXKgyV+cZz3Q5D/li6gYWsK4o8xaOTdXkDBCMNKq65ZxWF6USuu8goQvSdOPtOpOeMOHnIHtdZ75QzKU7db/9hDyAdbjgpyQPHQfiAt6JdNd4/hZWqVJyXri/CEvZ+UDxz1hBU1Z1aw8tnRJmt6kYnfWefc3+iZlGbUsLdf2NCqZxABsQyi2p8GHAIxK1hfn5QyP5PXnI9MMMRGPGZqpoFEgLqgZurw5MMqqBdBFuxxdydoUdEg+NCiZ5EtZ1ArL+f1GCKK7ldisW2rd/jhf6vAwFIoJkk5Bj8v8YRjpww8/fOGFFyoqKiZMmPCrX/3qzFc5fBjuvBMA4P33pbS0gxu/UgTCg0YMmzB1zAlv+HBzsNThuawgrX2SO3ERM6vYERbtEVcQA+TqVe2jK1+MtwejY1s3sWiOcK4oNyopkcjhcEydOrWxsREAHnvssW+//XbKlCnyU1a14tjenUMnFwHApv/d8OA7/6qoqJg+ffrw4cOzsrIqXEGfK+KKxEWMx2cYagNReXfaYpslOcaSp0EivLil1l1is1S6Q3L5K7mMzoIhGSyFBAmrGGpuQdpX1c3yLtElNssxT7ghFNt0yilIWMXQJTZLR+9tuXKpIGG9gglywja7R87HusDfLUtTEzIMUUEcYdEBgJzvy1Co16IrIFOEPc0f57+rde+s88qpS80RbnOta1+Dr7uur6CpknyLgaU8nHjcE2YpalZey/tYDtvlLdUoCp3wROSZQS1LO4KxfQ0+DOCN8bvqfazPO8NZPcdrz9IqA3Gh1OFJvgVLoSnZJhVDUQiZVKySoaZkm1iKYinUWR9YtQqOH4eNG7vrXzpw1NbWPvjgg2PHjtVqtcOGDbv77rsTO7y299VXX6HTdbQQiThvLIUuzbemaRTuKNcUjusVzGxbWvKfaZVKde+99/7jH/9Yvnx5Yo6mLb8ftm6FrVshFqMQmnTPHbNvvm7iFx9RCI2w6MZmGHL1ai2JrgaY+lCs0h1EAAyF6oKx70/fShID7Gv0H/eE9zf65XqE2+s8la5gcunaf/zjH3J0BQDhcPjpp59OPHWoyTd0clHI74tFwlf/7NZLLp23f/9++SlOlGr80cZQTJTw+ExjgBO8MZ5GSMBSorwWAIgYxwRxUpapwKiOCi3FRZU0JUhYnjRkKVTtj2w65QxwQpUvLE8dYoBjnvDUbJOSoeQckilZxo6mI+Qk9zAv2gzqKwanDzdr46K01eGOdUcxBZtBLUdXsmFmbecZZt2O9OeepWJoFUPJ1f2HmrU7672ihI3dWuiMpSizkg7wEgAYlExD9ckP3nvPYLFO/8l1ADoAyNQq8/WqU/6oQcmU2Kz/n73zDo+iWv/4O2173/RGDR1C6MXQUbAhoAIqCChefyLKvZarVxREEL0goqIiCopSREBEQHpN6J1AOqRsskm29zLt/P6YzbKGDgGiN5+Hh2dn9sw5ZyYzO+85532/L8ejTIO1zOVXiclUnSJZJWm854D+xUlAUT2CwaNGR+OrLMPfHDod6BomoW6FcePGmUymTz/99L777svJyZkwYcK2bdtOnToVKfdXi+zs7Hbt2t3NTv5tYDj+YIW9tV4hiIb4We5Qhb1jrFr354cUx4CsmXAVEfifBNN5HmbNAgB48klo1eoG2xVqwGrqTG0QJfnfo9ITOGq0IwSd4zUKEXnAYM2xuHEMa6EL3QwYQPcETabBVuL0HT99RpfSBCfI9jEqNYV///33ubm56enp2J+XBXA8ZPfnWT1FDr/X5Zwx/gmJXDFt8fLx/54ulYRcZkUEnqSSFNq8COCC3euhWQLHOB610isTayIBhWzibprNSNanx2pMPtrPcAAQLReXu/zCXJczyJyscuIYlm/1GFx+oRIMoNIT2HqxmuZCzR2osPVK0mnFVJHd2zpKKfTYz3KlTn8rvaKpRm4PMILfVfsYFY5jHI9uP4SwPtBgYN1ZxASekazPMtiqvMEqbxAA2kYr6/bHNMfiLvUwFI5JKcLqp/cePjt71iwmGJzzwczjx48nJiYCQJtoFcNDWqxKTOBAQEay/rzF3VQjF6TqoMbzQ/B5r8O+NXALDB48eOrUqYJQYdeuXRctWpSRkbF27dpx48bd6679DanyBgWhuJ6JWoWIFAbTJQ6fLu6S55PwpqnyBlViksQwq5/eW2bpmxJFCWYWy8L06QAA7dpdwcDKzIT16wEAjMbQnunTQaEAiwUAgGFACDkkSRDd7RCnBu4tChEpJoi20cpYMb569apqpyd10CNKEVGrTM4fa9RpvaKapALAH0u/Hjrj7VGjRv36669CgbFjxyYlJZWXlwOASqV6/fXXhf3xCnGZy6+iHR6LqTj3/McvjZ/25dLUps0AhSaoUlTSEoeP4ZGHZjEI2TQxYgwhhNW4xsoootob3F9mVUtIf83MltVPN9cqiuweYV0Px7Cu8Zp8m0dE4BnJOpOPzja5AECwrpJU0nJXACF0stKpkZCVnqCH5rrGawIcl2mweWhWQuKpuj+9ENtGKe/YJb/b/B2MxHqOmMBbR4VmKVUism6tqxKnL8/qoXCso0bUJ1lvM5a375nxwvSPAKCqqmrD3LnwzTfw3XdiAu+WoAkvasgoomu85vbTlnEc53LdonK0IJ0Spk4mhP8evPPOO5Ey0ILXamlp6b3r0d+ZZJW0Y6yaQ+hQhX1fmdXLcElKaa3El/k2r2BdZSTreyfr5RThCrJ7Sy3CDZxvvaaeyMmT8Omn8OmnsHp1aM/338MXX4DTCQAwcybI5SCXw5NP3omza6A+oxSRg5tEJynEDzzwwLPPPvvmK5NHd261ZuniWsV+27RZqgi9QSRq7U/Ll4etKwBYsWJFZmbm4sWLv/jii3PnznXs2FHYrxZTg5tED+rdo6ys7Ny5c5nb/hCV5z04aED37t2/+uornufVYiquRnoaAcIxWPHhuzFqpU6n++yzz4T9gmqDn+WqPEEA0EioJKXUz3CVHr8K44Xlvyjen6iUZCTp+iTr1WIqVSsPxwliAEZ3AAOkllAEjrXSKxQistztP2K0C9ZVglKSorqra3Z3mQYD645j8dPHKx0AICEJF80eNtqvm44AABBCn3zySfv27dPT07/99turFUtSShMUkk46iZLCRQS+/8evj+/dsW5R6PFoZDTCiy/C5MlXPnjqVMBxwHEYPx4AgGFCmzgO2dnX7eHSpUujo6PVanWXLl0uXLhw3fKR+Flub6lFcCwAADfN7i41Zzdk+bgSQnhaODjoivTv35+iqPj4+AkTJlRUVNytrv1NaKqRtYlScgj5WS5GJu4ar6kVxNRCJ2+mlWck68UETuFYz0QdiWMBi61o+RrD6nW23zaGyp08CX/8gTb/AQUFlw6Oi4OuXaFrV2jd+lqdqKqC3NwAy93Ij0MDf0V27dr1+eef79+/P3IniWOZmZn79u0TNoN+3+zZsyMLWPz05P8ulMoV67/90lZdNWDkaH27LkJmZQGKoqKjoydNmvTyyy8nJydHHivcxRKJpG3btrm5uSNHjNi/f//p06c/+OCDzz777HS10+DyhyONeATKRs0BwOFwTJ069eDBg0INwpqdUKZznLpLgkYvpWiGdXLAcxwAlHiYgyfPUAQuuOSafMEyZ0imBAEIkYP9U/R9k/VaiahPsl4pIo2egGBddYvXCnX/XW/6BgPrzmIPMIKQevsY1cDGUSoxWeUJHjVe38l96dKlr7/++rlz506fPv3CCy/8/vvvVyxG4liPRK2SCv0d//nKlC/fnFJVVgIAzZo169Onz7Xa4HlAKPRPILx5vV/5wsLCt6a9a7fbAeDEiRPPPf/8TU1BCdWXOn3HKx2CzGmA5dmGV8tlWCyWd999Nzk5ecSIEVcsIBaLp02blpWVZbPZfvzxx6ysrB49epjN5rvcz78KkQKhqGbe1M9yZTXihxY/Xe0L1jqKwLC0GFV4AlglJgc1jo4pLUod+2Ty6Md7vjQ+VG72bHjoIezhh9il3186eNQoOHoUjh6F8DBpyRLYsAHi4wGg6rHHz323HDVuDEeOMN//EBkQ08DfialTpw4aNOjVV1/t27fvO++8E/lVpLVUaxMBnK12iSTSFfM/XP7J7PfGjfS7nLgubtqHH2M4nprWCQBmzJghl8t5BPYaIdwr8scff3DcpZt/+67dZh9N4DiHkITEhcXuXkMflatCc7eCOzzLo10lFp5HgoV1qMJ+tMJu9TMMwjiO+++U535Z+IlMqTQQKqF1i58WBEubamR4yMCDi06f1c8IcetYjdkHAHY/DQjA40Fz51a8OzP7dN7f775v8MG6syhFpFZCxcjFwsqg4I9VK5/AFdm2bVvk5tatW4c+/HCQ5cORrgGW5xGqlVsjPT09Nzd397JlWobp16+ffM8eAACE4MSJUIm2bUFS0/qzz0KPHgAAWVnw9ddAEPDjj6Gv/pzNSoBlWZIMtX7O7Prvum0zJjxZmp+LYViHB4btKbX0SdHfYAIpGUUIqdMMLn+F288jaKqRdfzzukwDDMOMGjXKbDZv375dKr2yOk5GRkY4T9ngwYN//fXXtLS0BQsWhMfBq1atWrVqVeQhY8aMGTRo0B3qs9/v9/v9qF6aCCV+vjKImuEBMWIZhin0cnYGWinxAg/v51C0CFeRcMHHH66wtVUQGuo6C+hK/Krn6PL5uQgbFwHQPFI4HIKAlTm9E52QGCcWEwC+ps0K7hvQSCpXApQ5vV6Gk2K8zecWiagbVCu9O0RHR9/rLvyFqaqqEnSbBebMmTN58uSEhARhs3fv3kOHDt2yZQsAYBjWunXrxMREjUbzn//85+mnn+6VpK3yBru/99aogRkJCQlpHZo5aW74m691f+Bhn1gRHXTc16GNkG2w2hvslaTTS6krKta2/vMEatNGKXFycaHdKyHxjGQ9j9DW/HKFSt1/+JMbf1hMkmSfPn04hA6U23wMS+A4x/NSEvcxnI/hKAIP0uy8Vycd27392O7tADDyH1MEsWsvzXE8aqqRlzp9PKAEhdTo8fM8EjLkKEVkpsEq5C0w+2g/y28vMQ2mgsSbbyYBVLRPp9u3uJqGAgIw+YIxslCKKg4hu5+JujxXWz2jwcC6s5A4dl+SPnzDiwl8QKMoLOQae8leuZz27duvWbMmvJnWMT3LYLN5vNkbV7dslDxs5BMHKuw8Qn1T9JE5EwAgISHhGaMRFi68tIthoEYZBc6ehfbtQ587d3a3T1OKSGBZ+PprwHH3409eMZh2zZo1U6dONRqNGRkZP/74Y+PGjRMSE1EAzVz+6wcTxwx8fEy/4aNYnudvJoGgnCLSY9UHym08AqWITGuwrv4MQujZZ5/ds2fP8uXL+/Xrd4NHtW/fPjk5+ciRI+E9nTp1ivToAoCWLVsqFHcqIReO4ziO37n6bwcxF2D9gSIkbiuXlNB4VZCVEDiDiwK8P14mStPLMAwkHvq8zefByCTFtfQeLzgDha06lpwsIDGMDtIP92gLAMfnLqwacH+aXhallodSRQHwCJ2y+Fw031ssEZIwFwUxtxfFYzgAJPF0ld2MGAYAMK+vkcvaRid1MjSh1dTPa9jALcCybOSQAyE0Z86csMmFYdjvv/++cePG4uLi06dP//TTTwBgNBrHjRuXmprarVu3xmoZqGVP1njpKcUUAHRo2uhopd0q1W0rNolwwhlkFCKSxGBniaW1XpF8mZ7WqFGjMjMzv/nmG47jevfu/cEHH0iUcquf7hSnEX7z+zWK2nq2sOTsyQ4dOkyfPr1Dhw7CgTKK6JOsP2K0C3NUOIb1TdHbLWbThXyhwO7VP7475R9C7vNGaqleStkDTIkTdYpTN1bLPLSi2hvMNrsZnj9jcrprVgadQXpfmc3HcHsqrMJoLy1GfQ2FqiIfMtlt6XHqxmqZkO3N7Av2TNRdnky9XtFgYN1xag0nMAwOHjz43HPP5eXltWnT5ocffujatevlR73xxhvnz59fs2YNjuMTJkx4/rmJC1ZvaNK5Z0qfIXNen0yntFFGxUSJCdFtaKYJqiqN1bIOCGEAPEI7i02to1St/pwOs7S09JlnnqFpGgAyMzPHjx+/d+9eiVQGAa9CqZ6zeiOOExgAw6MKT6CV+EbfCm6aPVHlAAACx9w0e6rKmR6nbsijFuZf//rXqlWrFi5c+NRTT93UgRiGRf6at2zZUshNdndACCGErjbfdm/pIJXihLvA5sn2IoRoKUmkqcnpb79RXG3h3M5Zsz7o3bt3S6lUr5BGXTNnWYnTV+AMkBTZo0kjCUlkXQypELESSZumicmXqZwoxIzJzxxs0r6XP3jW5DL7aA2Bo8mT4fXXRAs+7bXgU6FY0xXfw4rvAYDs0ye4eXP9vIYN3AJJSUnt27c/e/ZseE9OTk5kAZIkhw8fDgBhjVAA4Hl+37593bp1u2KdiUpJN9AerbT7Gd4PvAjHCnZvtqT1kCjVJl/wcgMLw7Avv/zyww8/dLlcFEXpdDoA6NfokmZerFr5bEanZ08cu9RJi9vhpzmASm9QKSIFAwsDkJK4KiH+7NmzmzZtYhjmoYceitTeU4hIhYjUSUXCaoawGaeQyCkiSipSUGTrKCWOgdbjGlicfcbkpGwh2UVJ9plQhGNqKkRkgxaIosBCYyernByPqrxBQdNRU6eCR3eCBh+su00wGBwxYkReXh4A5OTkDB8+nGGusHYukUh+/vlnp9PpdDoXL15ssVjeeObx9YsXavTRM77/RRkVczpr76CWyUOHDLHZbLUPfu89KCiAggL4+GMAAIoKbRYUQMS7VkLiFI5fsHt/LK36uVHTwrHPcRyfe/Z0rcrOnj0rWFcCx48fB4DWeoVGTAEAjhMAgACipCJBKtfj8Sy5//6jMtme6Og9whrlZbjpkN9VU41sUONoKUmUOH2nqpw3eS3/tsyZM2fBggWzZs2afLUAhatw7ty5srKy7t2736GO/dVpF61UkLhgf6bFqKa+9OJ33323a+Nve/fuGTp0aHV1NQBc27oCgHiFRC8V3Zek10lFMoporr1kURXYPJE6jQJpNaFYu0ssZh+tlVAZyXri1Vfq+NwaqMfMnz8/UrBq4MCBVywWaWABQOfOnS8vE/41TlBIdJLQvRrkeEWbLhKlev/v6zYt/jxcOBAIrF69euHChUVFRQCgVqtjY2NvpMP2AJNn9QCG4RicqXaGnRQ5hI4ZHTxCKpXqqaeeevbZZ6+obFzLV0TYFBF422hlyKs9K0s55P77xj3Rfeo/QoVeeQUGDIABA2Dt2ssr1FJYj0QtgWFnTC5B7Dec7a0+U9/79/fDYDAIv+MCFRUVYR3ey1EoFIK8JEVRBEFsWvatx+UEAMTzy+d/yND09u3b33777dqHRUdDaiqkpoLwLGFYaDM1NVJrRykiM5J1OM8qR42jtmZmvz7ti7envvXic7Uq69q1q1wuF0kkKq0OAPr37y/UqYiQcccxrH20StB9mDJlSv6OHd38/lSL5eGHHy4rK7v8vEQ4LiLwJhpZx1i1nCL6pujlFCG9Mf+tvz3ff//9f/7zn9dff72WM6zA1q1bMQxbvny5sPn8888vX768tLTU4/Hs2rVr5MiRCQkJU6dOvbtd/stwqtrpYXkCAwA4We3MuVgS/srtdh86dOhGKhETeN8UvU5KAUC523/G7j+wbI114xZJv74+httvsNaysTCADjEqMRly2kqPVVM4BiRJ79l79OcN+5evDzRuCgCmxx7fv3z9mTWbPB9+WEen20B9YeDAgUuWLGndunVKSspbb7315ptvAsAff/zRvn17vV7/xBNPWCwWAPj4448fe+wxHMfVavXcuXMHDBgQWcnq1avj4+PFYnFGRkZxccnRSrvVT0tJAgBhGKbS6QvOnPji7amfLVgglPf7/b179x49evSUKVPatWtXy6/32mglVNtoJRc5GY5B5ziNSkxWeYPnzG5hH8cj7mY8QwQQwEWH92aPipaJwlNWjdSyv4QS6V+gi38zGjVq1Lhx4/Bmampq2NvxGuh0unemz5ix7BeFSm02lmM4Pu3bFYlNmwPAuXPnbrkzShGJ26oAAMOw7MNZmRt/dbvdtcrExcVt3LTpg+9Xz/l547iJzy9dupTh0Y6LJoPLHxbS4hE6WGFz0ywAbN++PXysz+fLzMy8vF0xifdN0YfVhmQUMaBxVGt9g9MJAMAnn3wCAPPmzYtMgPPyyy9fsfDbb7+dlZU1YMAAnU43fvz4Pn36HD16NCYm5u52+a/BKSHpLIF31Ylb6OQ0x7+96Kek5i2EbwmCuAU1fJufwTC86YiH9Q8PSWvXPFklpTnez/7JwGJ5dKjCFmR5YUH/UIXdy3AAQPbtw92XIe7fV6JWAkBUi2bQJ4PPuA9dW9Chgb8mEyZMyMnJKS0tnTNnDkmSZWVlI0eOPHfunM1mW7t27aRJkwBArVavX7/e7/c7HI6wZKhASUnJ2LFjhdF4VlbWz5nHKtwBhYjskaAhUCiCO7VDp4EjR6tUocSF69atC+fGCQaDM2fOvKkOt9Qp2kerLllYCFiE5BRJ4XhTjQwAOB4drLBlldvYm7SxWJ4v69V/95FzfkMFnDol7Dzz3U9MhRGqquBK2TwRwOEKu9VPK0QkjmHnza4Sp++mGr0nNPhg3W0oitq8efPLL7989uzZjh07fvXVV7XCdK8Iy6P7nnnBFWQ5n4cyFm3Zs2Po0xPe/3Htke1bmiqvvg7dpw+sXAnh+rdsASHa32IBuRwAjJ4AH52EON7jcqb17jvxnQ9aqq6wPtK3b1+J0VHpCTwz7UOJWnvAYA1yPAbAI9BLRSyPnEGGQ0i0agVkZ38uEgkRR2qAeQD3b98Op07BtGm1rwOOX2Pzf5lrW8xDhgyJHFU2a9Zs0aJFd75TfwcUFCkliXStSIJDgkaFYVgJhpo1SikvKtBoNPPnz2/evHm4MMMwr732mtHP2c3VXVo2++ijjzAMK7B59FKRPmINsUOMqpFaqhZTAIABdInXuINsZI5OHqFDFTZhZfC+JF222V3i9GUarH1T9FKS6J6gCa8c4Rj0TtIRGHaFRf8G/nYcPXo0EAiENyP1sUQ16wwIoW+//XbTpk2xsbHp6emRziS/Lln0cZ/+6XGqo0YHhxE2Q2mAY6Pjk/7x/n/BeFEoU8v5JDc3Nzo6unHjxtOmTRs2bNjVOmY0Gk+fPt28efMWLVqUu0MrgwSGIYAz1U6ViGR4/qjR0TNJe7zSYfbRGgmFEIqQX7g+FI73bhbH8rFSigAIWYdtUlOohPirHZLv4S1MSOzXEWAOVdhPVTnFBB6vuH5I/s1x/DiYzZCUdCka7DbA6mdAdb2iurpao9GIxbcerZCfn79ly5aoqKgRI0ZcI6PctSmweQyugDPISEkikHfieHFFkw6ddLHxMTJRBzXF87xGo7lOFZs2wSOPAAC43aBQVHmDhytsCEFrtWj7hl9juvUTyRQtdPJ20SoAYHl0zuxqG60UTJ8Ay+0rs3kZFgNAAFKS8LNclFTUK0m3bft2iyJWGR3b77UXdZuvrNfFlpebMOxG5uoa+Evj8/l8Pl+9TTjN8MjvcYcfFoZHFI65XC6ZTFYrpPfll19ev3nLZ5v3cRw375VJfTt3eP7tGTkWt5Qk7m8aTVwpEl6AR/CnRIUI8qzuKm/wviSdMIN1utrpZTjBoSRUaMUKMJmga1e47z4AsNlsIpGoIYrwpjCZTK+99trmzZtpmu7du/f8+fPbtm17Jxo6f/78119/7fF4Ro0aNXjw4OrqaiEd2c1y4cKF1q1bh22goUOH/vHHH7XKzJ49e1rN0FSr1QaDQZ8vNG3z6KOP/rZhg93PZJVbExSSTnHqAwcOlNvd0hYdm+nk7aNVAOByubp165afnw9/jn1RKpXFxcV6vf7yXq1atWr8+PE0TWMY9s2GbfoW7QDDkpVigysgqD/IKUJK4SYvjWMYj5BGQmUk6ajb8YUqLwdBHzUzU7j/LycQCFTYnOUs2T1BK6wMVnuDhTZvj0Tt7ecjqc3QobB1K4wbB8uW1UFtqIHrUVVVFQgEbvnwzZs3h0fFCnYIAAAgAElEQVQkaWlpHo/nlqviETpZ5ViXZ9xcVH2o3LYuz7jlQrWXZp1Op91uv/7xGzciAASA3G5ngHEFme0XTaVOH0LIHqAdAXrLhWqj2y+UPWtyrssz7i4x0xwXZLmdxeZ1ecb1+Ubhf2eAOVvtZDj+q6++AgClRjvuzXe/0mjYjh1R584oKQkBIIpCnTsL/5jq6oqKils+8Qb+Kni9XrPZfK97cS1u8GERfIEHjBi1Nqd8dXbpv7/4bl2ecUNBpdUXjCwWZLlq76UfB6M7sLmoyu6nhc0qT2DbRZOHZlmeD5fhEeIiNi/HarW63e6bO6v/bWiaTktL69GjR3FxscViGTNmTFRUlNForPOGCgsLIw3flStXlpeX33Jtq1evTkpKIghi0KBBV/x5TE1NjXxZv/HGG127do2JiRk1apTFYhHKuIJM5M3kCjKRNTgcjoULF06fPr3WOsn27dsvb45lWSG6EACatUtbm1uxPt9Y4fIjhLKrXevyjL/mVZp9wSDL/ZZfuS7P+Gue0Rm41FxRUdGBE6do5tKeyG+vit+PNm1CmzYhm+3qRfzV1dXXr6pOGDIEAaBx4+qksoYZrOtzmzNYvXr1inSe/fHHH8eOHXs7/Tljcl2wewFARhEZQmY0l+taM1jffw9LlwIA2GyQkwMAXM+eToYnMEy26EuqUyeLnz5YbouSirpHjKoFxxFhaYNH4AgyIgJnOJ7EMZZHwlStmMDbtm0bGXK8bNmycePGwdy58OabkJQEBkOoNpY1mUwNM1h/e+r5DBYAXOdhqaFHjx6CnFj/4aMmz/4Ew3ESx+5L0uki1gcRwN5SizPIdk/QxCskfpbbVWKhOZ7C8T4pugDLC6rWXeI1KZeFzV+Dhhmsm2XZsmXjx4/PyckR5DQ9Hk9iYuJzzz03f/78um1o1qxZ7777bnhz4MCBy5Ytu7UZrOty4sSJXr16RUZwb9u27f7777+8pNlHO4JMOMut0ROgOb6x+k9LJT179jx8+LDwWSwWFxcXx8fXXo8LBAJyuZznQ2t2/YePeveNf/Xv2Q0AEMC+UqstQEtJgsJxFx2aeBNmsHBAY8eOPV1w4d3vVmYf2Dvhgb6pzZtXegJHjI7mWpmwJHI7BAIBl8t1Z11LrVYQTnzMGNi1C0aNAkGojCRBq73lWhscX+444ftVIDJfwQ1i8gXD2TMQgLvm5uYQuqGsGqWlkJUFWVlQYwkRhw7pjh9RHzucU1Re5QkeKLexPIqSiQXryhVkeQQkjvVK1EXLRPYA46yxrqKkoiFNY+LkYleQzTLYaK72i+r6y5QNNPBXYOXKlVqtFgCiE5MwHAcAHkGQ4/0sF3bpxQCaauQIoSNGR4HNu6/MRnO8QkQyPL/fYBOsq/RYdYoIg4oKaMgRecf4/fffmzVrFhYrVygUAwYM2LBhQ503VEvR/o6+8l9++eVI6+rhhx8ePHjw5cU4hI5V2rNNrnybBwCMnsBRo/1UtdP751DW5cuX9+rVCwASExO/+eabsHUVYPnzZrdwQ0skkp59+496+TXBL3DP+tXHs0JJEjGAfo30TTQyP8sJ1lVLnTxGJnYEmKxy2/rff1+1apWl0ui0WroMenDjyTzBukIIKUX1XaoqRKtWEBMDMTGwaxcAwOrVoc2rrFreIA0G1h3n7bffDrt3tGnTZuTIkdcu/+mnnzZr1iwxMXHKlCl+v7/SEzhgsB0xOniEhBGzyUuTOJaskgZZPstg816mu1Obbt3gpZfgpZfgoYdCeyZN4v/vpcpxE6s1UQcrbByP5BTRTCMDgCpvYG+Z5WilHQHwgBju0rtEJxH1StKJCLxHojZeIZGSOIljc+bMUSqVQpkHH3zwwQcfvMXL1EAD9YmmTZtardY/jmePevk1DLBmWjlC6HCFY3eJ5WBE2JRaQnZJ0CBA58wuH8PGKsSDGkfHySUMx3MINdfJm2hksHUrJCVBUhJEODU3UIecP3++RYsWkXtatmxZXFxc5+mGxo0bF06rkJycPH369LqtPxJDzfS/wNNPPx2ppFVdXT116tShQ4fOmjmzo15G4th5s/tYpeOo0Y4QdI7TSEmi2nspq2bjJk1/27GHYZiLFy8+8MADws4Ay52scuTbPMcrHQjAFqD/MefzJ19+LePh4eFWIvuQVDMRS+F4I7WsZ5I2RiYmMKzcUA4AZmP59PFPmo3ljdM6H6qwI4Q6xWkaqe+FXq7BADNmwIwZ4HLdg9YjaIgivOMMGzbs+PHjmzZtio6OHjNmTNgcuSLDhw//7bffhM8LFy4kSfKjuZ8oxWSlJ3DYaBfhhCCn2ylWk6SSEBhW4vSdqXa2U13z7/jggyDYPZs2webNAADz5+MKBe4NesptAIBh4GW4gxW2RKXkdLULAMQ4znD83lKrm2FVIlJC4mYfDWRougzHsO4JGuFDnz59ioqKMjMzo6KiMjIycCEY8I034I03bu+yNdDAXaXE6TO6A+FV8gKbx+Kjg0o9BhhCyOQNttQr8qyeIIcwhhVedFXe4OEKe4JCIiEIQZrB7mcKbZ4qb8iQKnX6U5TShkndO43NZktPT4/co9FoEEIOh0MQxN+3b9++ffsiC/Tt27djx4630Navv/564sQJt9vdq1cvkiTNZrPTeUcUkh9//PHPPvtM+BwdHd2jR49wQyzL9uvXT1Cr3rNv/+kzZz775rtT9qDB5ccA2mjEakQfNniq/GwrlShRRvIITtsDTobvoBErMc7lckmlUnOAO++kmylJJ4EZXH6HL+hmeblGu2f96szNvwGAWCweNmxYuFEnw5+yBQBAReEuht9Xaumsl7RVEQCg6ddXLpd7vV5Tedmv33zxj/c/BgA1hauBdjrpy87spgkEAm63+8a9dIicHMX77wOA+8kn+RtbwCV/+gloGgAk779PnDzJDhgQfPVVAEByORfx91Wrby6lW4OBdTdIS0tLS0u7brEDBw6ErSuBXbt2iUk8I1mfabBWeYIAgGNY5zh1kkoCAJ3i1DKKaKKRBb2e8CFms3natGknTpxo27btrFmzkoUAjcuw+ukjRjvUxAMCgNlHm32hh6HM7ad55GFYDCBZJWmuVQj+WEV2r6BWFZlPNCYm5rrTcg00UJ/hESqye11B9kiFvUeiNt/mzbW4KRzrEKNSich8m7faG8yzegCAwLAAy+8ts7aOUhw1OniEKr0BjkdaCWUPMDTHn7e4AUAnFUXLRPlWz4FyW3+Gu8XI4QbqCJ7nWZattefWqsJxPJzcrFaddct7770XFxe3e/fu2NjYqVOnaiM8gc6cOSNYV2KpbNri5TZTlY9meMAAEAIo9zJxEkIvIqr9bJ6L5gHMAdZO80oKV1M4XyPShgAQoEIX01xFlXpZN8sDgE6MD2jV2PzEExRFTZgwIZyR0M+hU7YAhyBVKUqWk3lO2uhnT9kCPaKlOECjRo22bNny5ZdfapKa3P/8FAyAwjEHw+c46TZqUfhVcXNaDncXtmdP4QPS6QCAj4lh+/a9/WobDKx6hNlsrrWnU6dOACAm8GSl9HzQDQA6CZWovKT8IeQNDEYcMmLEiKysLAA4ceLEsWPHsrOzL8WPaLXQrRsAuFj+QKWN5VG7aFVTjWx3qcVD/+mXguNRpScAAG2iFK30SgDomai7YPe2aNACbeDvCI5hPePVWeW2Km9wZ4nFQ7MUjvVO1uskFABISGJniVlwTEyP0xwz2p1B5nCFXTiQ4xFF4H1Soi7YvefMLgDAAesSp1Z8+knL+Z+yPBKxNYP4Jk1CHzZsgKvkmGvgFtBqtbWmkRwOB4ZhYZfQ/v37h1JQ1CksywYCgZud1bhx3nrrrbfeeuvy/S1atCAIguM4lVYXnZjUpmuPXA8CQIKGjpPhy2m8wk8LBk2BiwYA2uV47YWnbWbTmDFjpk6dqlar1WqQyQLHKu1FLibsyUti2KOPPDr8scdqtagGaM4RIgJP1ckBoLsaTlc79VKRtmbRMCMjo2WnbgdrVgajZKJMg7XKz/p4GNQ4GgB4BIcqbDqp6BYEpcViMYZhV77ONhvYbCASQUoKlJSAICKYmyt8qTx6FISAmy5dIC7uhhojSQAQiUSiuvizNvhg1SMGDx7cJPwTDNC8eXNB1Puiw3fe4iZwTC4iLX76UIVd8G1neHS62slEqOiazWbBuhLIzc0VBjoAEOT4cy06oMNH4MgRhVqpk4iiZaIWOjmJY+kxf4ryEIRGeITi5GLBugIAEsda6hX1dvzRQAM3QllZmdFovHz/8uXLUxLin+zWzuOwCYONnkk6wbrys9yBchuPQEzgLI8KrJ6mNeFaBIbxCIkI/P7G0QGWE7J/EDjwgDINNsbpIk3VEosJdzhCzVRVhf7RdbBu0kCYtm3bFhQURO7Jz89v0qTJzSbMNhgMI0eObNKkyYMPPhj+5bwnIIRMJhNN0zzP//jjj6+++uqiRYvCPu9JSUnz5s0jSdJsLJ//6iRAgADhOBaeIsq3eTw0SwXdAYdVOGTV159nnz5VUVExb968b775RtiZqJQkKKXC+0OEOHt1pSnA/eerpat/+eXyLrWNVgrWFQBgAOmx6lqxsWoJpRSRgt+VnCKE1LSuIHvG5OIR2ldmqfYGix2+m5V9vxYmE3TpAqmpIDjGbdgAjzwCjzwCb74ZKjBhQmjPjWXBAgCYOBE++ghGjaqTDjbMYNUj5HL5gQMHFixYYDAY+vXrN2nSJAzDyt3+09VOAsd6JeqqvMFCm6faGzxR6egYp8kyWO0BBsOwJjVTWhqNJjo6OjwTJpPJkmrSkp+qdhrdAS/Ddo3XBljey7Behitx+rQS6pDRHtmNAMsDAI5hVd5goc0rPFT2AONnuYQ6l81toIG7gsfjGTFixI4dOwBg6NChv/76q0QSupnLysomTpzIMMyw515SaEI6QPlWjy5Ry3Bof5nVy3BJSmmnOPURo73aG3QFL4XxEhg2uHE0hkFmmdXHcKk+e7uinGKn73TG4BOdenWd9wmBY3DuXEgn5b//FcbH0LTpXb8Af2ceffTR9evX5+bmhmUadu/ePXHixJuqBCH02GOPCbllSkpKcnNz8/PzRaLrZP6+E2RnZz/55JN5eXkajSYtLS3sPbZ58+aNGzcCgM1my8nJadKkSWxs7Ny5c5MStEcr7RyPJCTeIUZ11OgAgIDLnn3+fPue95VfKIxLbjTuzXd9HtfONSsB4ODBg0KFJU5fucsPABgGAR7WLPpi2MR/dL//oc/ffDnjvvtuVlVHTOD9G0WFhT+VIlIQI71g9xpcfprjAaC1Xi5Ig150+JQiMlp2e5c3EIDiYgCAm4/NvypPPFFnVdUrJfe7JsV7s9y+kvvtwPDocIWtlV4ZLRPxCB2qsFd7g3KKEBG4PcAIWuq+CHHqHTt2PPXUUxaLRaVSLVmy5PHHHxfq8bNcpsHmodl4hcQVZLwMl6ySdonX8DzaU2p10QyOYXFykdETBACdhPKxnGBptYtWRsvEWQYbh/gBjaNVoisZ5WvXgtcLXbtCmzZXPIsGHaz/EeqtDtb06dMjc7HNnTs3nOtt27ZtQ4YMGfniq09N/bfP7fryP//8YPEyV5CNV0i6JWhOVTk5hLrFazEMTL5gVrkNEKhEpKtmVV0ppgY3jsqxuP0sl75vGz56NACcMtpIkmgbpcQxDH77DYYPBwDw+0Fy/SFKgw7WzcIwTOfOneVy+apVq5RK5ZQpU7Zv337mzJmbUqgym821ZBeys7OvnZ6SZdlbVnK/Bunp6adPn77iV4WFhc2bN3/ooYfCsu/NWrb66o+9HobHMEAIKAILh34DwMWc7PfHP5ma1unfC5eSFDXnpfEn9u587bXX5s2bV+L0napyYhiWqJTEysVHyi0A8O37/5GrVL9999XOnTsHDhx4mydS7Q0KYiUAgABwgG4J2kSl5ILde8bkEhH4kKYxNyLFflUdrLIyaNQIACAlBUpLgWEgGAQAOHwYBEmLvDwQ/jpSKdxASro6p74sETIMc//99xcVFZ08ebK0tFSv1/fr16+ysvJe9+uewbLsRx991LNnz2EPP4SKzwuWPo5hPRO18QqJl+HsAUZCEl0TNMINyiI4a3JxPBo8eHBFRUVRUVFVVVXYugIAKUn0TdarRGSlJ+BluCSlpEucBgOged7LsEJgYOsoJYFjAOBhuI6xahzDEIDZS2cZbAzPt9ApBOvqCib5q6/C+PGwZctdujoNNHCTFBYWXm2zW7dusXFxD4weG/T5Zj7/VLSEvC9ZLyJwm58OsHzneE23BK0Q1KEWi8QEDgAumtVIqA7RagBwB5mzJlebKGWsXJJtCoWFd4xVK0XkrhJLgK27sXUDV4GiqB07djRr1iw9PT0lJcVisezdu/dm7R6dThcZEqRWqxs3blzHHb0BEELnz5+/2repqanp6elbt24VNlU6/eS5X3oY3l1pmDHuCY5hGA6JCaxLtCzo8wLAyT07PS7nqcw9OdvWnz24P//E0XHjxr3yyisAoKBIisC7J2i6xmtSVNI9Py72ud2l+bm/ffdVjwGDO3fuEm7U4r/FFe0YuVguCpk1ghl1rNJ+ospxxuQiMKxbzcvrpjl/Htq0AYkEwkr3BgNIJKBUQnw8YBiEl4ZlMlAoQKG4J9YV1B8Da+XKlWfOnFm6dGnjxo31ev3ixYtpmp47d+697tdVWbFiRY8ePdLT0xcsWHAnZgFnzZr19ttvHz58eMuWLQ888EDYw4BDEKyZDg2w3MFyO8PxLIIT1kCR3ZtjdQOASCRq1qzZ5f4HPCCupqs8gLGy8r333vvXlJdReYEgRa0WU0ObxnSOU/dM1CYoJD0TtSSGVfuCDM/rpaI2UUoA4Hh0wGC76PgLZDJvoIEwkYGuzdulPTZqTHiTk8h/2bJDFxtPkMTYf0z+/ocfcsxumuNJAvMxHAaAAeRY3MVO374yS0JNiEmj39aohj2Y/t6bAJjHZIaSEkdeIapZnTdk5+WdOAclJW6PH5KT4amn4Kmn7tWv/P8CsbGxy5cvt9vtXq93+/bt1555uiIEQaxZs6ZZs2YAkJiY+Msvv9yTSUQMw4YMGXKNAqdPnw7HLWEYhuM457ROGtLHYbN6Pe6g31eScy5Zqy7ctdHv9ejjE4ZN/L+xY8e+/8Y/339uTLXF8s4nXwjrnlEy0QNNY8LJkme+/ur+JZ8xLtuEV/71xlc/FPhCItYXHb79ZdazppsWlEIA+8usriALNVHnCAAhKHX6CQwTNLRuts4Q+fmQmwvB4CVfRoQgGIRgEDweqDeLclB/fLCuJsVb57kO6oQdO3aMHTtWsKtOnz4tk8leeOGFWmVYHnloViMJ6djSHB9geZX4Ri/4+vXrw58DgcDWrVtbtGjB8OhAuc3mZ/RSkTPAsAg5g8x+g43jWC+HACD26resn+UEb5JEpcQZZI3uwB/7dn744RyOYxctWrRw4cLJkycDgIjAG9XkWBATOI5hgk1m9dN5Vk+qVn6wwmb20TTPN5aL8IL8mrNlAQCqq0EYe6nVUOP71UAD9YGRI0f+9NNPS5YsUWj14z+Yz0vELppViUizjz5YYZOoYtprZOfM7rYDhp50cm46KCUJBUVkGqxdEzQKisy3egQPYh/DqcSUK8gw+QUxWfvczVv0SNTEz5wPc+dGvtJTOrZNET4dOQLdusGKFffkrBu4Kbp3715UVORwONRqNXb1fN53mmXLlr3xxhs7duwoKyu7YgGGYRo1alRaWgosE+Wuvnj+KBMMlhXmvfXkQxzLNWuUNPnRQe+9/s9N23d26tK1/4hRjVRSBEBz/EGjwxVkm0tBCKijIiaQoqKiliz+BmqcScrdAQT2KKlImG2Kldd+s5Q6/VopJaxp0BxvcPnjFRIZRSAAluMpAi9x+Kx+GgA6x6olFCGsFQq2D4+Q4JJ1i2g00LQp0DRwHAjLXAQBgjb9/feDRALt28OePQAAdzS7zg1QX2aw7poUb53wxx9/RM5arV+/fvjw4XK5vGXLloKQFUJwqMK232AV7jCa47PKbfsNVjd9o7opta6GsHnB7rX56SipqHeSrnucAoAHAGeQ8bAIAFrrlTGXPQZhzphcgt9VtwRtRrKOD/g69h2U8UhIsffLL7+8VHT9epgxw/fDj5nlNobnk1VSYfyRY3HvLDGbfXQo/5TFDO3ahf6ZTAAAc+eGNhtURhuofzzzzDN79uxZ8cNSIQtCpsFa7PAJmQyaaeUtdIp4uQQA3DRLYJhKTFR7aQAotvu0EkoI9UAAPEJhJ3cAEBF4gkJyrVcxywKGAYbBunV3+PwaqBs0Gs0dta7Ky8vff//9f/3rXwcPHhTUDQUQAkF7XavVfvfddxcvXuzS5dI6XeR0WsuWLYuKiioqKkwm09OjRw0cMECY06o2lFoqy++vyajz8P2DnujbXUoSpS7/iUpHpsHqCrLRUipKdK2zk5JEn2S9SkRWuANnTC4cw7onaGsZWFY/faLKkVlmFYIEt100nzG5DlbYeIATlY69ZdYAy2sllJjA02NVjTSyWLlYGLfjGNZCpwCA45UOo+dWExsMGAAXLoDBADXZFSExEQwGMBhgyRIgSVCpoF8/6NcP7pHndJj6MoN1XSnesrKyWuZ8SkqKkPH+TkPTdDAYKTVV2/q5cOGC4NJRUFAwatSoM2fONGnSRC8mzT76QLmtY5S80BlwBlm9hCQ4NhjkOB4REUOHWpsCH330UWFh4ZkzZwiCmDJlSv/+/YPBYGMFxXPSRkpJUUH+4MGDZfqY2as24DgBABQGSTJi06ZNX375pc/ne+qpp8aPHx+ujfjpp/Tq6th26fFDBtLBIA6Al+Wu2LR934a1QgG1Wh0+R3LtWmLlSm/fAUzPwakaaQuNNE5KHjd5EEJehhMTeLdoucsfUATpq928PM8zf75iLMtefhnvCfcqWKGB+kMLJSWXSPJtnlPVTgBoG60UQsolVOgx5BASrCsSx9pGKwFASFhbYPM2+u0XRfFFAIg6fhgAMJPJ8dqbGtoPzz4LkybZt+7QznofAPb+vFFCEmmxaulVwj4a+OtitVorKytbtWp1C8cajcbOnTubTCYAsJKKYfomrfSKNlFKhOBYpb3cHegcpxZsEYIgdu3a9cUXX1y8eLFXr16dOnWaMmXK+fPn09PTFy1aRJJkOGaoS5cuq1evnjNnjtPpfPzxxyPTUStEZN8UfabBWubyA0CsXNwpWu6026/UtUtISDxZLT1vdgOAXkpdPn2ll4qaa+VFdm+mwUoROMPzANBKrzxR6TC4/HKK4BHSSKgHanzY/SxX6vSFVwbVYvJ4peNUtTNGJr5FN6w64euvYc0aaN0aIucX6pT6YmBdl507dy4VQp1rmDhx4kPh5Hp3EkHCLjJed9iwYQcOHPj55595nh88ePDx48fDX9E0vX//fq1WqwdoIsWK/ei4yQMAKgJSxbzb5bTS6GIAtZFjcgIDgGoaGQKonQKX/HkyUaFQbN26tby8XKlUajQaR42Ujh7A4wrOmDHDane89tWPgnUFAAyCXUXGcWPHuR12ANi/f7/f7x8zJuRoov/0U3FOTsyUKc4enRkEFAY9Onda9MXnCCGVTo8Y+q233go3oWYYAkCGQ4oEi4GgwxEkAWQ4eDkAgCDHZ5ucJhppMbz17j04Bh4O4kY9IbZafC+95H/iCQBAajUX1v4BAACWZZ1O583K0twJ7o5R3kA9J1YuLrB7EAISxwQ3lFPVzmKHX0ISQYZFGAYAGAY9E3V6aejBF/R7kjZviM3cE65HZLeJ5odcRUs//qRaf07QD1X36V3sCjgoIkMkkd/VM2vgzjJjxozZs2ezLJuSkrJ+/fpr/54wDLNy5crCwsKePXsKb6uVK1cK1hUAHPhjwwNjxuUBIAAvzZa7A3KKiFyFUKlU77zzTngzUuOwFiNHjkxKSqqoqOjRowdF/Sm/MoljJI4DcAAgJQn8BibnSpy+82Y3gWFiEjf76GOVjq4JmlqHdYhRETiWb/UEOR7HgEdwvNLBIySniIxkvYwihKaFwlKS6JGoxTCI+fxTyMlJ7t8fe3y0giLrzLoaMOBWjioshD17wOO5fslbpb4YWNeV4p04ceLN6prUIZfLNCxfvvyrr75iGEav1w8bNuz3338X9lMUNXDgQOGp0/F8dYnFx3AA0D5eJ4wDLGY37fPk+LCMZJ09wBQ5nDiGSVSay0cJABB3FfFZr9c78Z2ZKS1aAcC2Vcv6PTpSLFfwYtmwif+3fP6HQplCk72MF3eJ02BYSJ3WDSQoNCeN9o6x6hSVdOvWrbtPnHUro9soqZaJMbBvH7jdAABWKwDI3a4u2ccBgBeJDrbu4uWCajHVPlp5pNJRGeTFBG7hyIst05pq5GcqbLEEAQCypk1lNZlQa8GyLIZhDcZNA/UBB8OfNdsQgiipyOKnswzWDtGqEodPShJeqwlX64ViCEGl3R0t00PI/PIBgDcpxZXaCgDEVrPYZkVisa9xUxlJYBi4WT789upoKms5bjzD82KyxrH93/+GOXMAACZMgMmT7/Y5N1AXHD9+/P333xc+l5WVTZ48ee3atVcrjBB65JFHtm3bJmy+++67M2fOjFzpu3D+7M5vF4z417v5Vg8AyCgiI1kvwmDp0qXHjh1r3br1Cy+8ILkBXQ+E0Lhx45YvXw4AUql0zZo14akHmuMzDTZnkImWiTw0V+L0IcQnXtOqqXAHTlY5CQzrkahViclMg63c7aeqsfTY2srmYRcuCscxDBMCZjvHawTrqhahF9z27bBnD0gkSRMmXPe8rk9KCjAMQOgFVw+53W517tz5hRdeeOqpp66dw/i6tG3b9uzZs5F7bk2K926iUoUE0BcvXszz/Pbt25OSkubOnSvEoTA8n2Ww+RhOISI9NHvEaO+VqI2SidtGKxFAgc2zr8zK8gjHsB6JtVe4r8tTTz9dJY8FgF8WfrLmq0+3rGP4FJ8AACAASURBVPhh5rK1Kp3u+N4dQgGSono9+kRg+458mailTsm63BQAW1yC7dylc3qL1Fp8QG+EwK2KBQzkwon83/+FMwwAABw/Do88AgAOgjAcy43Ravok6ygC752kO2CwKkSEiMMr3IEKdwAASLy+OPPdK+rqQWjgToAQmjlz5pIlSwiCeGHylDaPjuEQdIhRNdfKz5hcF+zebLO7S7zG6A741XoACPp9boc9Kj6xyEMrnT6tmCp1+jDAEKCi2f/1KiSFNm+bz/7b6usFTOMm8rwcoZU2CB1o2+bo1H+XZx9vduhQ2tk/SxlduBD6UOMi08Bf7qm5ePFi5GaxoHJ5FU6dOhW2rgBg3rx577333tixY5csWSIsemi12ikvPO+XUoInluAk/vzzzy9ZskQ4ZOPGjYI07rU5ePCgYF0BgN/vf/311x966CEeoQKb1xGknUEmVi7uEq8psHnKXYFSV4CSYtfIGqMWk4JXu/BW6hSnyjTYL88iWGDznDO7SRyLk0vK3X4AIDCMQ+hEpSM8g3U3uAXTatEiEP6OmZkAAAZDSPkdx+Gjj+q0c7dtYOn1+pdeeun1118fPXr0Cy+8EM6CebPUiRTvvSI2NnbKlCkej8disZw+ffqRRx4hSSrLYLMHmGiZqFei7oLDe87s3ldmdV7MfSCtVbvEREeANvloAGgXrYyTiwFg69atJ0+ebNmy5fDhw/Er2SscQj6GU4pIANi7Z8+BU+f6PDpi9cJPMAx7YMy4netWBnzeisI8giAnfzjfVJjTP1mX0PNJ4Vhhyjhp03rYtD4RwNwrI6vVagAADLrGa5OUEofDQXo8VwxK5jju9ZFDX5r0/MC33wIAnYTqk6JXUGSlNyBIBqvFFLFgAfi8/8vp1erqQWjgTrB48eIZM2YIn/MtzpYIEpSS5lo5QhBkOQBIUEqSVVKa5ys8gaDf98HzTxWePfX+srWtOnUtc/qapES1j1adt7h1UnHvRB2GAYFhwuhdRFx6VMsM5QdtztQXX80/fWLWq5NWj3sOCEpKYEnffQ0A8OijIdmePn3u8unXW/5CT01ubu4//vGPkydPkiQZzvF87ST3tRY9KIrCcVxI17Fjxw6n0zlo8OBSlqpy+eUUQfPogt0LLPP999+HD9m5c2f4nXgN7H/2qRI2z5ndRXZvjFzURCNtE6U6XGG3+ulUrZzluGjsWr6wQs5Noydw1uRqqpEdMzoRQkJQV3Z29tSpU/Pz84eMeOLByW+SONYrSVdSI9kjowiNhDK4/JkGa58UvZSMsLG++SYkAVpeDgBw/jx8/jkAQEwMjB597bOre1auDJlWAlVVIAhCYVidG1iAbpuSkpL33nsvJSUFANLS0r788kuHw3GzldA03b59+x49ehQXF1ssljFjxuj1+vLy8tvv3u1TVVUVCASuUSAnJyfSQ+vtt99GCJU4vPvLLAzHu4NMZnbeC9PnrDlvWJtb8eL0Obvyy9blGYV/GwurnAHm3//+d/jwZ555huO4TZs2ffbZZ8eOHROa4HiUZbBuLKyyB2iE0P0PPbzqzMW1uRW9HxzWrG2HX86XrcutWH6isEW7tD2F5evyjDuKTTxNo1DK9Nr/3H36Ca2frnYihI4cOaLT6WQACoDHhwzhR49GAPmNGysAFACCZsPQoUMjT9nsDW4oqFyXZ9xUWLUuz3i4wsbz17qGDMNUVFTc3t+hvlMnD8JfHa/Xazab71Xr27Zte+WVV2bPnm0ymSL3j474EW/btecv2SXr8oz5Vs+RCtu6POPWC9U+hkUff4zefPPgoqUde2UIJTP69DlSbrV6gwghd5DZVFidZbD6GRYhZPfT+S/9EwGgNm3CrXz44YcSmXzW8vXr8ow/ny1Zl2fcU2qhA8HQc7d27Q2ehdVqdbvddXdV6jV/iaeG47jIwCapVNqzZ8+ZM2d6vd5rv6SeeeYZ4RAMwz777LNa356udgq3n93jtfvp3wur1uUZHxn3fLghHMcNBsN1u+f3+yM97v/7+UKa4wMst6PYtC7PuLfUsqfUsi7PuKfUTHN8MBisrKy8Yj0X7V6O5xFCzgCzqahqXZ7xt4LKdXnGE5UOHiG/3x+p3Tr5vdkWXxAhdKbaueVC9ZlqpzPA8AgdM9p3FpuDLPenqpXKK7+M0tOve3bCCVZXV99IyRviuedQejpKT0cxMQgAyWShzc6d66yJGurAwBLgOG7r1q2PP/64SCSSyWQTJkw4dOjQTdVQVVX19NNPazQamUw2ePDg7OzsuurbbXJdA2vBggWRNmuHDh3CX3E82nqhem1O+eAnn27XrdeqMxfDptXHqzdP+Pf0dXnG3/ONE96agUXMWj344IPhx/KLL74QqjpndgkGmckbXHHorFDJylNF/R574vuD2evyjGtzKz78eeOlFwZCpQUlmw6c3XTgrLN5CwSAXnkFVVdXXCjbeCzv13yj8M/g8tfKM1/aty8CqOrUKXLn7Nmzwydl9dGCdVVo8/gYdttF07o844XPF6GuXdEjj1zxEv0vGFgCt/8g/KW5hwZWOIUtADRt2tTpdIa/EpKmkxT15hdL+j325NfLlgs38Lo846aiKi/NospK1KgRAkCzZpWUlCxZsmT9+vXMsmXBzl2q+g82eYM0y+0uMa/LM67Pryx2eDcWVv163nC0uBLRdLiVb1etAQBNVMyq0xfX5Rl/OVfmY1gUHuc0GFhXp54/NVVVVbUmJs6dO4cQYhjm2gYWz/Nbtmz54osvwkNlhBDLsmfPnjUaja4g8/Ph7Oat22AYNmjQoKLyyv1llkXffhdewRDG6jeCxWKZMWPGpEmTfv7t998LqvaUWgQba8uFauE+31ViplkOIXQ1AyvX4l6XZzxYbuN4nufR3lKLcOCOiyZh7CxkaYzEbrcLx0aaUzxCNMfVrr1pUxQdjaKjEUUhACSRhDYHDryRs6tjAyvMP/+JAFDXrnVfcw11ZmAJFBUVTZo0KfwH6Nu378WLF+u2ibvPdQ2sWsvko0ePjvy2wGj6+Uzx2tyKZ/71nxWnCsMG1psLlxIkNfE/M9flVKzLMw6fdGW/V41Gw9fMDuWYXeHDlx84/X/vzlqbWyFsfrjqd+HDt/tPlVQYEUIX7F5hCFLq9LlatUEAxpenlrn8YbvKUPN51KSXIlss6NEDAaChQ+fNm5eYmKjVal999VU64kVCc9zuEnO+NfQO8DPcjmKT/f1ZCAAlJ1/xEv3vGFgCf8sH4Ua4hwZWrZWU5cuXh79iGObFF19s373X6uzStTnl2YbqH/YcFZ6X1acKEc9fGlLPmnWpxnnzEIA/JnZDQaXZG8y3esJP37o8Y5bBui7PeKY6ZMYJsxGvfPDxrBW/rcszrj5bIkweMDSD+vZFffuifftu8ET+Bw0sgXr71HAcJ3jWCsTExPj9fnQDBtbl5OXlCVXhOD569OhIwa0RI0aEy6xcufLEiRPXropl2enTpzdr1qx9+/ZLly4VdjIcv69mvqrU6fu15nY9YLAKs1NXM7ACDLfjokmwsYR7e12ecX2+cV2e8Wy1EyHkdDrDbsfCGIa/9rLFFenfHwGgSZNu6qC/roFVN77JgUBgxYoVAwYMSE1N3bx587Rp04qLi3fv3h0MBp9++uk6aaI+M2jQoGnTpgkr7j179qylPh+nkMx95XkmGBz+wssSqRwAEEIA0H3QkAUbd7fr3htwzGup3vbzT1esvFO/wVZfaMm8uU4hqhncPJDeet67bwkxSgihoD+0EK7RR3mB5BEqdvgIHOuVqEtRSeUUAQBehgMekRjWJU6TpJQkKSXd4rUEhg0eeGkGKy4uTv/JJ7BnD3z88WuvvVZeXm6z2RYsWBAZ+kvheN8UvaAXBwASEh/QKCqsWf+/zP/4g3BvqeUoHblJkuT48ePvv6+n8cgeHMcLPJwyPgkAMDrYaN/uQ9M/+H/2zjvOhuv948+U23vZerfqbaPsLhaLEDYkyg+JiC6JEkJIvlpElFQiBPEVEr5aiN47weptiWWx2LW93N7vnXJ+f8y6rlWyapT7fvljZvbMmbnjzp1nznmez+f2bunpsGYNrFkDR4+W7YvjNIuO5BnTS62+IkExidcKkglI/JrJcb7EymXKCwi8Y8/+NeMbIqetebhMI+IbXN4jhRZ6/19w4EAg9ep+PP93DY7jGzdubNiwIUEQr7322ubNmytS3HdPPv300+vXrwMAy7KrVq1CfoLVviGi6tWr9+jRo8Gdcwh3M2vWrMmTJ1+/fv3ChQsffPDB3r17AYDEsaYR6iAx3+iiTheaEYCYR8j5ZJHDc6LAzPodjmGRyX1bMhfHoG6oXC4gC+1uTu80QiZqFRMkJPFMk+PvEqtcLl+zZg1n11izZs21a9c+ghwr5y+8ePHi3r1727ii9ZcaDD2ecU9aWtrvv/++YsUKq9Xatm3bQYMGvf322+StxH6DwRAWFub1PqJV5HNCcXHx3TINd+NwOGw22z2FFX7+daE6Plki48pcEW4odImVAlGZI43bYugWX+vIoYPLli3zz3AEAF2lKj9vO8AjiKYRarmAty9b76RorpqJh+MkDi6alQlIm4cGAGNx4foFcz+YMBXH8cQwVbCYb/PSZSo+hw97TWZ3TKw8rraXYbnMXKPROG/evFKjqVlSYx6Pt2nTpuDg4KFDh3L5EBXl99/hzBkAgHPn4NgxkEqhd++yP02bBrdqkmmaLikp8SnjvXy8CjdCRXA6nU6nU6vVPvtD79u3r2PHjk6nEwBSUlK2bt3qu/5btmzp3Lkzy7IA8M3ydTUSkgCgOC9n9aSxfx0+cO/uYmMhLAyOHgWJJLvzOywCAMj4bLxHIgUMAEGwmB8XLD+SZ+Kq0wUE3jRCnWGwexi2aYSah2M0i47lG7kyYeJhHkVGo5HP5/8rRnjPmBf9rqFpuri4+KFcpWvUqHHlyhXfKpcyjxPEiGlzlHxi4rCBAMCw6FB28cldW84d2p+cnDxo0CDiXi6Wb775pn+V4ujRo3/44QduudDuPpZvAgAch7axIQRAaq7B6qVra2WxMr7RaAwJDT2cazC5qSSdOkjMpxg2Nc/ooBiVgFfi9ACAiCTaVgoiMMzmpVNzDZFyUVxQ2fCV3W5/tC/n8ePHXUlJrwMsBBgIMGTIkHnz5lVkR7fbbbVag5+47w1CgFCZ0cLT4XGrCBs0aBAeHj506NCPPvooOjq63F81Gk3zV+a9TSKRSCT30BS0eKjKrTt6Gbbs/xIwVhNGl5b4AqyYsDASx19//fWgoCD/ACs4OPj3eXNjghUXS21H8owkjrtpznoW6WSiPJuLYkFIEi6KwQAQgCooaPzwj2N1mpOFpkyjXSfT+jQSoVkzPgC3wkVXbrd76sKlJ4+eOrpj8+yZMHTYsLcHjqgWon646AoAdu2CNWtur9rt8N//li1PmgSvwEOCI3Aj/Ou0bt06MzNz//79YWFhLVu29H8m/fzzz1x0JRCKSEGZ8ktVm3XtqeP37gsAsrKAK8J3OGJW/I/bdnnICEwqRQj4BF7i9F4stYVIBDctTgAIkQiUQl6jcCVXhAUAJI4l6dQ4BhWRdnw1eQXvmu7du0+ZMoVblsvlP//88+zZs4UqbZM3OxAEkW1xRspEh3P1JgohTdifq1evWrXq8uXLs7mauzupV6+ef4BVr149bqHU6T1ZaEYAfBzzsuhkgalZhDo5UnPZaK+qltAUBQAYQLhUyBlxJoQqrxjtZjclIPASp0fKJwHA7qVPFpgbhatkfLJVdJCQvD3ZJZVKPQxLYJhPJtRFM5xx7YM/+7Fjx74HEAIIAd4DkG/fDh7Pv+lm8zRDq7IjPOYI1oYNGzp06EA+rzJfT4QKjmDdExbBnqwSB8VUV0ujFaJvf5xZv9N7xK3LJReQCIHNS+tkwoZhKhqx33/zzeTJkxmGCQ0LW7dhY5NGDQEg0+i4UFpmZn54y4amb3dmKErGI7wEwbAg5ZOVxKSVonl8QRW1REDgpU6vSsjzLyC/m82793oiqgPAz/8ZdmzXtmHfzmzRqVth9g2l/mbPnu8/xCecNAn27wcAyMuDrCwQCG6LNWzeDLd0Yl/6EaxX4UaoCM9mBItF6FShOVouDpUKAIBm0alCc3W1VC3iMSzC/GIaikUkjv3f/3XZtHGDQCT+YsGy2olJHpM+8+ThJsERb/ToeO8D6HQQGwv5+ZCVxfJ4xrj6DqtFqlAc+2WxV6lSi3hNderDeUYSw0pdXgGBAwZumlWLeC0iNdyhPTR73eyoqZU9wu/3qzOC9aLfNY8wgsUwzOzZszdv3hwcHDx27FifQVyRw3M834QQkvFJq5e+cenC5H7v2q0WABCLxQ6H4+6uHA5Hv3791q9fz+PxPvnkk2nTpmEYZnZTB3MNDIvqBsujFeIjeUaDy6sV85tHagAAIaAor9Fo5GZaskzOtJIyfW8xSThpRsonkyPVOGCcc251tZSzivKH2rnrWlq6N7Zy7Xc7kTjmpJjUXIOUTybpVL77DgFcMdirqCRcEMYgdNXosFy72CQpCSHUCmAf1664uCJ+zE9rBOvp87gB1qvA4wRYAGB0eYud3poa6dmzZ+Pj40fOmNfsrc7cn2popZUU4tRco81L8wiMhxMtotRWg76gsNCrjXTQqFmkmk/g2Rbn2SILALgcjqkf9oisUm3w5Glc1SGOgVLAixARShLEMvmhXINcQEbLRaES4d3+hv6kpqaO/Xb6pzN+AYDLZ0/VTkzSF+Z/2bsr63aWlJQ8itfptGkwZgxERsK9HOBf+gArAMezCbCK7J6j+UZObFot4h/JMxpd3hCJoHG46qjfrBwn9ismSb3FvHLh/BP7dk1ZutZYmF+pStUaQQqRvkS/cLGUT8QoxLftyd98E1q3hoYNoXlzmDEDPv/cHRyy/VDaiT07Gr7RjrstGoarImTCCyWWTJNTQODJkRrA4K+bpQwLUj7xRkwQxSBuRobTMn3YT/fqBFgvOo8QYD2AApv7eIEJAHiIeb9RHS66AoDY2NhyAqf+uFwugiB8OkEsQicLzEFifmWVBABoFh3PN0UrRJFykZNijuQZ62gkmNMaGhpq8VDH8k0IgYtmAKBhuKrQ7q4TJOP0qzw0+3eptW6w/O4XdbZDR3zrFldomLFdB9l/fzmWb3RQTIRMmBim8j03Lhvsl/Q2jYjfJEKNY3A831Ts8FRWSU5tXj1lypQ6JSXbOVmslz3AetUFuJ84Ho+HK5fzbVGL+DU1UgCwWCzvjxzb7K3OdotZX1gAAJf19lyrOyFUQeIYxSAnRafmGkVKNRMSk+/welmWZtE1k4MzLuA7zSKJZMLCFbnXr93ISOc6l/BJo5u6YHKfMXkP5RqcFOOh2RMFZu5GfQDNmjWLVklnfT6MIMjaiUlmfcmXvbuW5OXYbDaGYZ7a5QkQ4AkQKhXUD1EwCB3PNx3KMRhdXrWI3zBcRSPkYdhih+dEgclNs6m5RpObctI0wxN2GTziq2kzslN3S+RyFwtLN2z5bs5M7YQxkVO+hM8+u911s2bw+efQvDm6ZT5oNZmsRkOjNu18D48LJVaD25tv9wBAlFwkF5ByPtksUoNjYPcyx/JNXHQVKhFUUor/hasT4AWEQejGLcVOCiO++mkOtyyRSH55oBWxSCSy2Wz9+vWLjo5u1qzZqZMnG+tUlW+F9SSONYtUR8pFAFDi9Ni89Kliq4UGi4dKzTU6KcZFM9yc4Jkic6xC7FMHFZB4YpjyntMg3Ju7qKhQlnpgX3apg2Ii5aLEcJX/W3kVlaSsyCPPyEVXcgFZg6A/bNkyZ//+7b/9VtYuOxuuXYNr18DpfNQr91wTCLAeAqObSi+9XfhQ7PBcNtzhEzl37lyNRhMREVG7du0Mf9sZAACo1yip04AhNrNpUr93Br2eYMq5AQDppbYThWaaRZFykU4mtHvpvdn6HKuL86XSu7x/l1hxgMY6VbPYsA3zfxZLZV8tWlm59msOixkAHF5GzicRgINmnRSjEPJMborAsH98b8YwbMWKFV9+X+ZTK1dpqteLB4BBgwaRJAkdO0LbtuA3wf/P8HggEsFzbG0U4GUiVil+LVjOIGTxUAoBj8sr58aT5AKyyO7ZlVVidlNBYn7zSE3TCDUOSBxbM7h+klwbkpb61zdD+s6ePfu3n3/yTz+/MOyzhSQ/Ly8PAZwtMl+1eZBAwAqF0lsz3TwMAwAXzaTmGDmP0Ty720ExAGByUSwCAodih8fqpUMkgkZ+MyYBAjwABqFjeaYSp0cp5CWGKXEMq9zsjeOXb+zatevmzZvt2rV78O59+vRZsmRJTk7OkSNHmjdv7jOTLkeMQlw3WE6zKMOJUnONXoYFgCAxv3W09rVgOcOio/lGq6dMpJ6iqJEjRwYFBUVHR8+cORMAID0dzp6Fs2fhlmsw4XErLl3QZF2rF6Io90X3K2b0Fjs8Mj7ZLEIj+P03qFoVqla9XQjVqFHZlgMHHu3SPecEAqyKggDOFJqvGu3nii0AUOzwHM83XdLbfJWuGRkZI0aM4CbLMzIyPvzww3I9qMTChGCpKe1wSovkFStWDHijSZhEgAA5KSZUKkgIUyaGKQW33hh0MqGTYtKKLQAQoxSHSARKpXJEz276vByBUGQ3lL4RG1wvRMEiZKcY35fb4qZwDGsUfm9zQ4TQ77//3rFjxwEDBlxITz9bbHUKZGIeUV0hwHBsxPQ5yzfvKLuX9u2DPXvKbA0qyMiR4HSCX4FMgABPD5pFhXY3t2zz0gZXWa2ZgMAbh6swAIZFPBxrHK4icEwj4qdtWQ0AqqCQgqzrPwwbQHk8AHDkyBEAAAzTt259onotQ7OWglYdOr3Xc++lGzctrpt9P9qXkXv01GUcJ0iWFuA4dWtkmkUoXCaMkotcFJOaa7iot/1dYiVwTOCXWR+IrQJUEJuXNrq9SiEvOUIdKRc11qkwDKMkirZt22o0mn/c3T/V3ev1Tp069X4tK6sklRUiFoGXYdUifpCYn6RTEzhWRSWpH6IIkQi40SwAmD59+qxZs/R6fU5OzqhRo7Zs2QJt20J8PMTHw6FDXBtJbk6rLilJPTrlj5lAe6lyx8Ix8L298AicfGDKysvKi5pd+OzBABrrVKm5hhtmp4tmSxweBqF6IQrVLf2ny5cvc5VKHJcuXbq7k0rB6k8HDwQAl8s1cvSYmu26BekiAcDg9JrcVJbZ6WFYHIAFyDQ6bpicDELhuFeHl/03JSQkhNncGQZb+4a1hSQBADTLXtLbEAAOGAsIAMQ8QiO6tyTVjBkz/nMr1+TUhYtTl2/kxskkPEIhcZ8qNIXGJWCvvG1zgOcfTpjK4PKqRXydVHih1Hoi35QUoQoWCyiWPVVoRgA4BhSLThdZGoUrGRZFJzbj9g2Prdyh78D1C+YAQFJSErdxsFK5bt++3qdPdK4XP3bBchtBMm5n81oxqblGAMAxaBARdLHU6rl9fwMGEB+qJHAsy+y8YrATOCYgcCfFhEgELoopdnhOFJgbhSsDg1gB/hGlgJccqZHyCB6BA0CoRJAcqVYKeHD8OEybBgCwZk1BcfHcuXOLi4vbt29fzgNRLpf72xHa7Xa4DxYPlWv3+Jab6NS+uCdWKY71m9FO9XfrA0hNTe1wnz75ZlPMj98eGfhxUpVw37edReh4vqnI4ZELSBLHjS7vkTxj0959yFatAABOn4YhQwAAdu8GlQoAymw6XzoCT9OHQMYnkyM1QhIvtLu56Mo/xyI5OVl5ayoBAN5+++17dsLpdnbq9m6NN7sG6SIPbV63Zt5MikWHcgw3LU4RSbwRGyQX8ACAQSgz7WRStZjQ0NBBgwZxeV06mbB1dBAXXbloJsvsZBHIeQQLiPtu2730oTwj5Rfq+Vi27LaWafrpk/qzh7noCgAiZMKmfLblrzOw8eNh/HigKACA9euBW9206fGvXoCKU1JS0rt3b7VaLZVKU1JSLl68+KQavxwYXF6j26sW8ZtGqKuqJfVCFAxCWWYnxSLOZF0r4reJDeZUE4/lGVPzjMpQ3YVjqZP6v+ty2HuOGtdl0CfBwcElJSVGs+VkgUkTpgOA5T99m3v9KkGQADDjP5/oC/KaR6pDJAIWwckCk4NiuDfyMKlQyifzbe5ThWbZrTd+DMBJMaESQZJOlRypkfPJQrs7y1yBzJJ334WICBg79ilerwDPB1ardfXq1atXr7ZareX+pBLyeH4JTxoRn8AxyM+HDRtgwwaz0dioUaPvvvtu0aJF3bp1K2fO5nttBgAMw967j30yl3flZdhwAfjmBEud95YZa9SoUfnVw4eha1eIjLydB0KSEBmJIqNcUdFyEd//XeKq0cFFV8mRmmYRai4f6xIhhoQESEgAn7Fj3bplWxSKB124F5ZAgPVweBiWS30FALP7jkFRrVa7Z8+edu3a1apVa/jw4fPnz797d4vFkpiYOHLkSFog4aKrOeM+XTV7uvHyeRYhEseaR2lcNOukyibCq9RLbPNuL5ZlFyxYsH79em6j72ssIHC5gKcV8a0UgwM0DFepRDwRSVjc1IkC891Hj4iI8F8Nlgi46Kps1WUlvv8evvsOvvuuLMDavr1sdefOR7lYAR4JiqLatm177dq1s2fP3rx5U6PRtGzZklNAfszGLw0hEkFTnZrLuwKASkpxE506MUwJABgGWhG/SYRawiO4fCyEYQ6KCZEItC7DhWOHvxnYy+10xFSvPdZgaP/LLwc+HJxnc781bHStBgkjps2JrFz20//x1zOyC0vEPEInE/nuOAahcJmwUbgqOVIt5ZN5Nhc3MxgiEVAs4uH4a8FyHMMEJJ4cqamhkVaqSAlhaSnk54PpH6pSArzo5Obm1qlTp3v37t27d69Tp07ewyRgbNu2zb/9f31agwAAMG7cuDlz5tStW7dRo0bLly9PSUm5ZydWD00xbGWFKEaIVVFJuBjL6ik/tefr88MPPxSJbISF7AAAIABJREFURCqVaurUqV27doVKlWDtWsjJgTfeKGtUrRrk5GA5N/lZWXERQf67V1NLqqgkyZEaAYFz+ViVVZJa2vKKDy89gSnCh6DE6TmWZ2IQqqGRZluc2RYngWN1g2/bMyUkJGzfvr3cXizLXrp0SalURkRErFmzhiu4/WvDn8biwgsnjrAMAwBVFMKaEWqlgGen6GP5RppFcUHybyZNfPODYYMm/4Dj+K5VSzMzM8v1jGNYo3Clk2bPFRjChIROKuTjGB/HLxpsNTUyANC7vFqf1ijAjz/+mJ6enpOTAwAajUav19/RnVAIcXFly+npgBBERJSN394ZmQV4qvzxxx/nz5+/dOlSTEwMACxYsECn002fPr2cBdMjNH6ZCJYIACA/P//kyZOVKlWqW7cuAOAYNNGpV69atWDH9rCwsBEjRjSPDCdx3EnRIh7x28mTAJBx5uSXvbv2GzepW3Rs5I1rs9f/mThzbqHT+80fm7lRX1NpqSooSKpQMvnW3D/X51KAEpM4lSAcg8pKCY6BiCRqB8lO5Js4NyqtmH+m0JxjdZ0sNLeK1gKAgMRfwcdJgAcwc+bM3Nxcbjk3N3fWrFk//vjjvZseOFDm1HRrNDpx//7xAACwBCAfgEvMctFMod3DzaIMGzas78DBJjcVJb9vmVGkXCTjk2IcGY1uAKiikmjFfKXg3vkkAoFg4cKFCxcurMhHu1sSCMew1/yejOSdD0qoUaNMj9rP3PDl5OnZHL40cGbPLEJ7skrWXS64bnIghKweatu1ovVXCkwu7wP2zcrKqlGjBgBgGNa/f/8lS5b4X3wMwzQazaxZs3ztTS7vlsyiTKMdITR9+vS6TVv8kXateceuAoFg9+7do0aN6tmz58qVK33t3TSz60bJjmtFRQZjntXFuTvvyyqlGDbTaF93uSC9xOp/PuXG1bZu3Xrv8xaLEQD67bfHuGx38KqZPT8OXbp0qVy5sv+Wzp07V6pU6fEbPwOettnzn3/+2aRJk/r168+YMWPNmjU+dbqhQ4dyDfwzfMPCwsxms2/fyZMnc9uVmqCftx001I1HAMuDgrj7mvs3e/N+sUw2YtqcdZcL0j8dgwDsEVF7s0sYlr1isK27XLDpaqHVQ3EdppdaSxwebplF6Fyx5cG/Bnfw3nuoUyfUqRPSahEAiokpW/3ss1fW7PmF46HMngcNGuT/2zt48OD7Nh079rb7+J3/kgDkcnlqaipCaF926brLBRl6Gyp7HhWvu1xQ6vQ8+DTuZ/b8EBw8iDp3RgCoVq3H6qfCPC2z56dPYASromAATSPUpU4v94rA5WPZvPSDTY5Hjx59+fJlAEAILV68uG3btomJiadOnQIAkUi0devWpKSkkydPXrhwIS4uDgCUQl6b2CCulnDUqFEkSS4eP1yKwZ9//tmnT5+ioiIAWLFiRUFBwahRowCAj+NKAZlnc6eZvLWDhBiGMSyiEcqyOC+UWAkcC76znHDPnj3+qzt27Hjrrbee7IUK8JhcvHixmi9HAQAAqlevvmnTJpfLJbpLBeOhGr/ofPPNNxMmTOCW09LSlEqlx1OWsfvLL78MGjQoLi7O/527sLCwbdu2R44c4bTCg4KCMAz7EiHCUOru1o4vkQDAm6G6kh+/D6NpT2R0VrceQrlCrtJcv/h3eGyl6hgGADgGLaO0OIZVU0sxDLN6aCFJ2L20lE/W1soAwMuwNIvEPOKOd/R/ZOtW8E9Gzs6G7GwAgAYNYPz4x7hIAZ5TBg8evGzZMs4rUywWDx48+L5N1WqIjQUAcDiA01yIjQUAj8fz7eefx/Xpw41g1QtWHMkzXNLbvAybZ3O7aSZWKfafsnhaNG8ODRuCywX3ckgM4E8gwLoHc+bMWbx4MUmSQ4cO7du3r2+7iCT8B2BlfNKX4no/sjg7s1sUFBSkpqZu2bLFYDCkpKTY7fbKlSsXFhYqNUFf/29V7zeayYV8hkXn9Rabl0kIU3766acDhgwrcriPbFrDRVccCxcu5AIsDIPEcBWTU1rook8VmgGAxDG7l+aiqySdKkh8x/1Wp06ddevW+VbjfHOCd8IwDAGwb9++xHfekb/0o7jPGUaj0WegwaFUKhFCZrP57pjpHxtfvHixXNp77dq1Y7mf76eA0+l0uVzOpyAbeO7cuYkTJ/pvsdls/qs5OTnbt28vV0J18uTJJUuW9OjR47fffhsxYgQAfMn96rmc4HICgObCOc2FcwBgqJ+wp2r1qnUbzFy6RsYXIgSEywUAwLLWrJucF5tOownjk0dySu0U2zBYKuXhFItOlTooFjUKlggfaE5VDn7nzpjbDQD4wYNYaSmqVIlt0AAA2JgYl8tF0zT+PNXzisUBxdTHYsuWLUOHDnU6nTqdrmPHjsOHD+dmNu7Nf/5TZi2wbh106wYAKyZO1Fss7dq1a+n3NqUW8ZIjNYfzjNdMDgCIUYjrhTyrVHGhEITCZ3SsF5lAgFWeJUuWDB8+nFvu16+fVqtNSEh45N7ee++906dPc8sqtaZjx44CgaBbt24AQLOo49tvRdWp1+M/X5UU5IVVrbn3WkGdcO1Vo4OTgMuyOHVSYWquwcOw2tg7qlj9TQMwgBgpr9jNcGnysQpRpskJAFoRP0gsAICsrKy5c+fq9fpOnTqNGTMmPT19/fr1OI7379+/S5cuJSUl5SwIfv311088HgyAXrmy1oULJ0+evN9YCLpT7AcFtH+ePy5durR27Vr/LRiGhYSEPKXDuVyupxRg7dixg72zNjYuLu7cuXPccu3atRcsWLB58+a7d8zKynI6nTNmzOBWMwB4AAKhKAoQ4XZ75QqPNggQcuoiq7xWvzD7xv/N+CZ8zw7f7qL8PFHlsnhUf/UqqFQiDBlZdKLYVktOXncwdhopeRjtdjsf5tvv5ATnABSdO/NKSz3NmtlvnaHTZOLz+YEA6yUAIbRixYq9e/f+8ccfFEUBQH5+/qlTpx4UXd2L/v37UwCjR4/evHmzfw47iWM+rSkxj/hXfn4RAMMin9wDiwAh9GCjtleHQIBVnm3btpVbfZwAa9SoUWKxeO3atcEhob2++NoqVXsYVkDgHoY9nGus88bbETXrRFSqemLvjoOb1rXo1NVfKd7moVNNBg/DRivE9ao2eeedd9asWQMA1WrHTZ8+3dfM7qXTjB4WITGPcFJMpslJ4MDHiWKH52yRWYd5GjVqVFpaCgBLly6dN2/e2rVrzWYzy7LDhw8PDQ1lWbZ9+/Z//vmnz/ts+vTpvsKS9PT0HTt2dOnS5e6P5qSYo/nG14LlwWIBANi89PF8U0KYUvXAOdMA/4hKpbLc0krmMJvNGIb5i4BUvPE777zzzjvvPL2zLcfT8yIsVzceERFx4MCB9evXp6amxsTE9O3bt0HDxgqN1mIoK90IjoiyW8yIpnr16qXVaiWSsoK+1wCkcsWk/63uP3m8+vyZnE7d/v5iKgAwNI0DFhZT6QHncMkBDWNVjbTERb3tisF+3kIDlBUtPrqOIo8HAEKhUHjrouE4HvAifDkYM2aM/281R1paGsuyjxBAe73er7/+2hdg5VpcF/RWN82Gy4QlDs8lvY1F6JmWVrhcUFCQobcVBYU3i9byCZwTNPEwbNPHuSNeIp6jl6TnhFq1avmv1q5d+x93+e9//5ucnNymTZutW7eW+xOGYUOGDNm3b99vS5biQkm2xbnrRklaseVwrtHioRo0bS5TqKxGY6M32kVUqQq3dKLDpAIRSeTZXB6GjZALix3us8XWP1evPnPmzNZ9B6av382Prs615AzX3AyrE5Nc5SAACHCiaYRazCNuWlx7L2Ry0RXHggULAECpVK5atWrFihXckMD27dtjY2OrVKkydepUlmWFd4793m/4qsTpsXroY/mmUqfX6qEP5RhsXtonrh3gkaldu/bVq1f9t1y5ciU2Nvae/xEP1fh5Zv78+TExMUqlcuDAgS5ubg4AAFiW5dwR3n777XHjxnFZ7cnJyefOnVMoFP3791+0aNHEiRPVwSFTlq79aePelp3eAQBdbOUf1+5YsO/4riMnuNGCr7/+mscrC/1fq1s3Lq6O1+0EvzFXxLLcWMCl4Z/fXLcF9u41vt8bADzBIcYtO0xbdxz932oDX1ji9AJANZXEZ7oQFyx/rGeJTgeVK1fE8jbAi4Xdbr+njWDz5s0rGl117Lj85581AL43Xt8grsVDnSo2u2k2Ui5qHK6qrpYCwGWDXe+6t67VU+HgQahSpVbj+g6DaeXR84OHDd+SdrnA7vYXM3rFCYxg3cHJkyd3794tEAg8Hg+O47179x40aJDBYHjALgsXLvz444+55b/++is1NdUnD+2PhEckR6oP5hg9DMPJD2pEfBsRxGfYjGOHvDFVKtd+zdfY5KZufz0RRuJ4rtXFIlStVlxBnpFmWPLW/Skk8XCZ0O3xVpORYqlALeQROK4S8uQCsnmk5niBCXkd/qcRFFSmVsKl3vvQ6/V6vX7ixIkqleqbb7559913vV4vALRp06ZNmzb3/OAxCjHDovMl1qP5RgLDvAxb5ZVUOnnidOzYccOGDRkZGTVr1gQAu92+f//+AQMGPH7j55YDBw4M4ZSdARYuXKhUKqdNmwYA8+fPHzdunNlsbt68+R9//NGwYcO2bdsSBDFixIhyFiISoSBIpSAVmk++n6UOCXm7z0cSpQoArDjGIERgWJcuXdLT0//66y+FQvHGG28o1BpSdUfCCsnnAwAOYK1a8zyOSSLUmhMnAIDmC47XrE8xiKmC6oUoouQiimUP5xk9DCsiCRfNHM0zJkdp5P+Ujnlfli9/xB0DPAdcuXIlMzMzMTHRN0TKMXbs2JkzZ3K/oj6USmXLli3nzZtX0d55vJgGDaRRUcacHAAgSXLsLUFaOZ8XLBaUODwmN3XD7LxksGEA4TLhs0hyv4uMM8ervN62xYcjGJEYvO7mlaO4nMUAGEKBSLMMq9UaGxtrNBq51aSkpKNHjwJAcXGxUqn0FYSXo127djv9dDjHjBnz/fff3+8QJjf1182yKQwCMAZQZZXEaSgpABHnUVOan1eYff21pi0AQCcTGlxeN81GyUVGN2X30hgAAqihkbpzMktLS5OSkrgMdKvVyrIsNytULhGKpukuXbps2bIFAHg83tixYydPnoxh2J49e1JSUu7+3+/UqdPGjRuvXbt28ODBkJCQN998k6vAuh/cXAl3to3CVfdrRtN0SUlJeHj4A7oKwEFRVHx8vEQiWblypUwm++STT3bv3n3+/HmdTgcAO3fubNeu3bJly3r16vWPjZ89jzZF6F8eCABKpbJnz54dO3Z88803fV9RX/ktAJAkefjw4XKThiyCPdcKHOztr7/TZl3309eFNzLT0tLq1KmzYMGC+Ph4383yd25JtsFGCgVenGRvHSVEIlALyQyDw2u31p/9XfWlSzxR0dt2HwOAmhppTa2MRehgjsHkpoLE/CY69UW97ZrJISDx16O0Yt6TKaoyGo2BKcLnH5Zl33///T///BMAgoODd+zYERMTM2nSpDNnzshkMn9/QI5BgwbdU336fni93k6dOnEPF7Va3bdv3379+r322u33cM4eN8fqAs64KUz5ABEs/26NRmNoaGjFz+S+7NwJ7doBQJhY8t2+E3KVGrHs+umTViyqkHpWxXG73VarNfgFHOUNjGDd5vz5877oCgCOHz9OUZRvWuF+VKtWzT/AKlcz74+XYc8WWQCAi5MYQAoBz2sxFoCIRYgAcNpsQboIza2vkVrIr6WVpeYacqyumhpphsGOAOR8ctLHH65ZsxoAtFrtjh07yqWIlXc1J8nvvvvu9Pm/PW63saR46tSpISEhfT8a1KJV6z/++GPu3LkFBQX+pY5cSVqVKlWqVKny4A8OADYvfdNSls5c5PCUOr3lihYDPAI8Hm/Pnj2fffZZ/fr1vV5v06ZNDxw4cL+A6aEaP7c0btzYf9VsNv/yyy+Xcgq+/H3ltGEfuJ0OANBUrvF5r49mfjaUoSmapletWuULsEpLS48ePRoaGsqjWdBEAYYDgMvhmPJBj8y/07g2Z8+e7dChg++rbvFQN1wsTyatpJRk6MsSHzHAih2e40eP7N+y8eKpY2FXMr7v9H+q+CTS6SCdjgKbJUpRVcIjdDIRgWGcS+5rwXICw+wUzblXBXh12LFjBxddAUBJScnnn38OAH/99dfdLStVqrRq1arExMSKd+6mmdXrN/qeLHyp3GR3+EdXAFBcVHRq/6GQhGQAEJLEs/vtnTEDvv4aAIAucxzJQiyR0gQQHJu3uHXfQRTD8h6movYl5mW7CqWlpYMHD05MTOzTp49PNreC1KxZ079YJi4u7h+jKwD46quvkpOTAQDDsN69e/fp0+eezSiW5fKu5AIe/9aXz+KhLptcNE1P++SDozu2jH3vbXNxIV8gqKQS62SiSioxp7ZVTSW5ZnYCAA/HrV46onFLnCAAQK/Xjx49+h/PcNvO3RN+Wzll2Xp1SCgA7Dpw6FCu4XCesdu73Q8fPnzt2rVPP/1UIBAQBNGzZ88xY8b885UCAACbl07NNbhplnNif7CzVYCHIiQkZPny5SaTyeFw7N69u06dOr4/cYM63PDVPzZ+UWjduvWsWbPuqI3FsDY9B9Rt0nzsvMUCoahV1/c+nPhdYqsU30x6VFQUt7Bv375KlSp17ty5Y7d3r1E8LroCAJFErKt0x0tCYWFhNqc1BSDlk411qhCxgPNKJ3GMh+MIEA4otm5CSGT0zSsZ1sTGBZOmX3u3V/LK39s3q9fwvY6HcgwOiqmmliRHanx5V7WDZA3DVYGM3lcNf90cAMjPz79ndAUAvXv3fqjoimLY1FyjsEZ8eEwlAAiNipmydF1y78Fcdbnv6P1HfKat15hlmKvnz7poJjXX6KaZR/ooD4nLBWYzmM0+ITehy8WzWnk2a0HGRWWojnM8fBZn8tzzsgVYXbp0+fXXX0+fPr1s2bKUlBSGeYgvnFarXb16NWfY16BBA9/byYNRq9WHDh3Kzs4uKipaunTp3RNqBQUFPXv2TGjYKK+kBACsHtrDsGFSAQYYAEgUylP7d0dVqd6kXYdh3/70w7ABTSLUDi+Tb3NxFYUsQllWF8WwNTTSN2K1rNvZ7K3Ow74tq/Eu4WToHkhEWOiNSxfComOnLF0Xl9TsvbFT3DSrEvK4JwSO4zNnzrTZbHa7ffny5cIKq5tYPJSHZjlPq1ilmHO2KufPGCBABRkxYkRxcbFv3BQh9O3AXm6TPq5xs5827xn69Qwcx9fN+rYoJwsAWrdu7dPF/vzzz+12uyooZOqyddowHQA47dbCm1kA2LDvZjVJ6eA7RHh4OCcAdsNOHbxpsHmpHKsLAxDguIxH6mRCAEAIu3T6xNYlCwGgXtMWfKHotSAZVxjLw3E3w3DebX62tmDz0v7Blc1LP8XLFOC5oUOHDr6UVgAYMGDAPWfGcRy/nzng/eARuFbMJ0XiKcvWx7d8Y8rSdZrQMAkw/sNCqzdvGzDxOwzD54wdMeH9zlnnT9u99JE807PI+GnfHubNg3nzYNgwbsOVLyZfHz0645NPOg54nysqfzDFDo9v6gMAcq2ugpe0OuqlCrD0ev3hw4d9qxkZGRkZGQ/Vw1tvvZWbm+t2u8+cOfNQUiXR0dH3nCFGCHXs2PGPP/74+8zpEe1brlswGwBhGJjcNAIkpF1mfUnO1YztyxdlnD5RrW78kKk/XjHYix0euYCsoZECgJgkpDyimlpaSysTkUTjUFlpfu61C2XyPx999NE/nlv37u8WnU49tGV9WHTspMWrhVL53ZJ0PB6v4qEVR4RM1DJa6zOcqqKSvB6traqugLttgAD3YcGCBb5hY7vVMm3ERzgGoVGVAMMSw5W//zJn2dHzfx48Ubt27cmTJ1+5cgUhxHlrmg2l19PPsyxy2qxTBvQY1bHVjUvpGAa9R08geTyCIBISErZu3crn81kEFoo1e6ibFpeIJBCAh2VtFJ1tcZIY1iBceWXnutKCPABY+fO04mN7q6jLcqGEJN4iShsmveM2uWF27s0qvWkpK3vM0Nv2ZesDtbSvAsHBwYcPH+7Xr1/37t2XLl06ZsyYJUuWqFQqACD8JM5Zli3nn/FgGIb57rvvBndse+HgHlVQ8Pj5SzWhYWAufTc54Q7RQZd9z+oVc8aOOLRlPcPQR/9cEiUX1QmSLVywICYmRiaT9e/fnyu/ffI0aABDhsCQIXDLBSRm5LDKP/xQc/ZsceXKSRGq5Cg1//5ThBTLniwwnS2yZFucAJBrdZ0uNJ8qMLvpl3DQ66VKcqcoKiIiwjeoIxaLCwoKFIrHFbd9cJL7gyksLAwPD8cwTB0SaigqBID+Yye93W8gAAhpV0rNmNFjx/00fRoACISi8b8urdOoKdzy4fEVYjDsHbpt129kzZ0zu7S0tEOHDt27d4c7k9zvx+WsmxleHgKMxLE2sUGiZ5syEkhyf0V4TB0shBCfz6dv5Xa07/3BB1+UeQuqBDw3w7po5uCmNXPGjYyqWj3v2tUOHTps3LgRw7DIKtVzr1+Nb9HaYtBn/p325Zdf7tmzl9SEZGWkU3brrl276tWrx/VjtVpphr1kZ0ucHpLA6VtzGTwcaxapUQl5CKGdO3dmZma+v3On9u+/uX3AZgOSBJ9A67VrnJJ1scNzPN/EIFQ/ROGmmQyDnYdjTSM16sdQgwskub8o0DRdXFzsn/Locrlu3Ljx008/LVq0yLdx7dq1Xbt2rWCfU6ZM+eqrrwAgNCrmp837BEIRjmGtY7TlXEPMZnPTpk0vXboEAHK5fN++fQkJCUeOHElOTvY904cPH/7zzz+X6/9pJLmDyQQPfPqUo8jhOZ5vQghFK8Q3LU7AIDFMFSG77xv+i5vk/lKNYPF4vOXLl3PDtnK5/H//+9/jR1ePiVarDQ0NHfDF1B/X746uXjMoPKJhm3bcn6wM3q3PgPFjRl+5cmXq1Kk//TSjaZMyfQcegflHVOVUcStXip05c+by5cu56MoHArhhdvrqoVgEN8xObsXupbNYEQJMzCNoFj272foAAR4GDMN8s4QtO78zYNxkQIhyOwHA5KFcNJN17uSccSO7D/vsxw27m7TvyEVXgyZPm7Z2R+O27V0lBe1fb75hw4YpU6YcOHx44oiPg+TSkpKSFi1azLylnA4AOAZJOpWIJGi/TBGNiM/5imIY1q5du+HDh2sBID8f8vOB8+Sh6bLV/Hy4JUcUIhE0ClfhGJZWbHki0VWAFxqRSFS7du3p06e3bt0aAHg83siRI+8p1Hw/Nm3aBAAhkdGTl6wRCEWUw8YidDjXaL9z6lmpVJ46dWrVqlULFy7MyMjgSp2OHj3qP2Jy5MgR3zJC6Pz58+XUeR6XiAgYMAAGDICHHH0IlQiSdCocw7IrEF290LxsVYRt2rTJy8vLzc0NDw9/2lqLXq937969ZrO5TZs2/pPx/vB4vJUrVy7ddUCu1kxdsZH2uBWaIABwWC0SueKdzybMX7qiUVytAwcPvjV0tIXBZXyCTxAGl/dQjqFFlOahBAyvGOyX9LYih7txuAoAO1loKrC5XTSjkwoP5hgYhGKV4rrB8mP5pmKH51CusVW0NiC2G+B5Y8mSJV27ds3Pz3+987sA8NPIQWajccrSMrefvAtpiGVFYgmOEyOmzQEEtRs2afNuT7Ne33/c5LRD+5vViElq087ioQ7lGA6lXbpwIb1uk+bnjx4aNWpU48aNfRp1F0qtLprh6nkxDIQkVuTwnCk0x4cpMQC7l/YySPnBB/rGTYNEfGz3bti7F4KDS4cO14oFGADwb1dshUoFETIhVy1fSSUJRFcB1Gr13r17S0tLRSLRw45EVq9e/er1G1OWrtWG6favW2W9cm7YNzPyHFRqrrF1jNZ/6k0sFpd7x27cuDGG3Z6VatKkCbdQWlqakpKSlpYGAG+88Ya/IfpjUacO/P77o+1Ks4gFBAAIAfXyZsQTkyZN+rfP4QlDEIRara5IAWAFcTgcQqGwXPa6w+Fo3rz5jz/+uH79+oULF7Zp0+Z+818xMTFvtWx2JTOTr9QKxRIAKMnLGfvu28biooTX2xRkZw3t17tt/4/rN2/lsFnPrlvS++038+1um5cucngqKStkAebxeBBCQQppscNjcFFmN1XgcBfY3FI+WT9EnmVx6l1eEscahqs8DHvVaKdZJOeTsRXr/InA6XHLZAEZ0pcciqIoinoc6zqdTjd8+PBu3brJEXVw++YDu3aMnbdYIpMjFmEYpoyItphM748cAwCAUNKbb1eu/ZrZUDpvwmcNW6fUTky6abLZ+LJss4NBcP740V6ff9FxwOCi3Js3r2QkJCSIYqoTiM2yUzdtHpLAWAQ4YAiAZTEAsHhohkUSHpGaa8y2OG2VqqZXjbPEN4wozsMOHXKG6fZ/+zPZrKmmdUvwU+K+bLBfMzlIHAPADE6vhE8qBI/14+NyuQiC4PMDiifPO9zPGidGeDcGgyEzM1Oj0TzUwygpKWnfnj1mq604L3fel5+fP3fuWtqpHr16K4W8cOk/DPNER0eHhISkpaXRNP3ee+/NnDmT+xaNHz+eGxgDgBs3bmg0mhYtWlT8lJ44eTb3qUITAFRVS8xuusjuFvMI5f3fTGia9ng85aRcXwheqinCZ8mKFSt8Ls5ms3ny5MkPaEyS5P81TfBpPS/+9it9Yf7mxfMn9ulqvPK3x+NZN/9np80qkclJXZV8m8vhpTEMqj1kwriAwJMjNQoBr8jh4aKr5Ei1kCTqBMkj5SKaRak5Bk5YIVYpbh6l+eceAwT4N8jIyEhJSWn3ZsryRb9PWbYuKDziwMbV/+maQlNehSbok+9mYoDNHffpuSMHufabF/165sDeSf27mw2lMTVqAwAC7HLaqegataKr1wSE3HYHn8+v3vLNKwZ7usnjYZGQxJvq1FoRPznQgqnyAAAgAElEQVRSHSTm827lO5I4npprdNFMhEwUF6yQ8IhCuzvf5gYABiGFgCw35HvVaL+kt3H5W43ClRiGnSk059lcEODVZvTo0TExMU2bNo2OjuYEqytIeHj4tm3b/pw7Y94XoxDLAsCBAwekttK4oHuHceUYPHhwfn6+0+lctmyZb/CsnGJRXl5exc/niUMxbFqxBQASw1RxQfLGOhWGYedLrJ6XcRzrZZsifGbYb0mAcNhstvu1hFuSu1YvTeIYzaJxc3/b9etMc3HBBx98QBDE3Llzi3KyR3ZsNXnJ2rhmr58psuAYNAxX/eP7yt3wcVzCJyweCgAkPIKzS8MAEsKUDIu4Utgouah+yL+cmhYgwAP4z3/+U1hYCAAet8vlsB/etumXLz5jGWb5jO/7jZ0IALnXr9aKb9SgeSur0SBRKHqOGltakHd0x+bfp074bNavAMCybI36iQDgsFkkMsXw6bMliDJhAiGJx6kEAgyEUpmYR3CvGU2EaopFRQ53WpElw2ADgBiFuH6oAgAwDAOA/Jgq0L4jq4twUMyFUptWLPA5mmtEfBFJNNKp1EIeAC9Jp0ortsj5gVnCV5ozZ874PJ71ev3HH3987ty5iu8ukUgkEomvBpDP53PFiY/M+++/v2HDBm6ZJMlOnTo9Tm//yI0bNxYtWuR2u3v06BEfH1/urzwCb6JTuWmWE0YJlQga61Q4BoKXUZv0paoifErcs4qwqKioXr16xcXFAIDj+Lp16zp37ny/Hk4XmnOsLgmPSI7UXDc7Mo0OAYE3j9IggOsmx9o5P/44fZrX6x33868JKR0AQCvmn167dPGiRTiODxkyZPDgwQ8+Q66KUKFQniw05dvcUj6JY2D10GFSYaNwJY5hdi/NvZcDgJhHNI/UPClbjwoSqCJ8RXjMKkKOunXr/s2V7wFI5QqH3YZYtnGb9qNm/hfD8OK8nLDoWAAgAB3/YyFfoa77VlcWoX2r/2jUpp1crWFpGr81ob958XzK5er68UgAEJB480gNcjvvWXLroJgDN/Xca3STCHWoRAAARpf3YK6R+5HEMEAIagfJqqvvyKphEcL9dLHKrT4CgSrCF4W7qwg5Nm3a5P840Gg0er3+oXpetGjRoEGDaJrGcfynn34aMWLEY57qunXrli5dyufzBw4cGBcXFxoa6qIZ/1rycquPzPXr1+Pj4y0WCwCQJLlr165WrVo9Zp+BKsJXjtDQ0DNnzkyePHnkyJGpqakPiK4AQCHgiXlEs0iNmEfEBcmra6RCkhAQ+LkiS5bZiXSVGRb1+Ww8F13xCVzv9OYw/EsZGX///feQIUPWr19fkVO6bLBx0VXzSHVypEYuIAvt7kt6uy+6ilWKI2QiJ8UczisLtgIEeA7p37+/b9lhsyKWrdOoKRddzR4zfOw77e3GUgBgAatWP+HAprWnt23AAW/TvZdcrYmSkGmHy6YOacrbsf/gN9/pwa0yLKJYRLEo38Vws34A4KKZbLMry+I8lGPwMCznlXsi31Ts8NAsulBqQwhx41gIgUxAVlWVj3vKhVOPGV0FeAlo1aqVf9TVu3fvh+1hwIAB2dnZu3btunHjRsWjK6uHvmK8PbVicHlvmMv0PLt27bpp06Y1a9Zw2VdXjY49WaU+443zJda92fonIhO9ePFiLroCAJqmf/nll8fv88UlMEX46Oh0uokTJ1akZVW1JFYpvu2toZVVV0tJHKsqQhcybiS0Spm2dntM9doAoJMK4oIV685kNO/Qxe10/PrVGADYtWtXRQp9q6ilDoqpEyTjbNGSIzUXSqzV1ZJMk4OLruqHKLjxyjybq8Dmrqx68XIGA7wKDBw4UC6Xb9++XaPRVKtWbcKECVZj6cWTx/7auCZ16waSJFtU1l11IqvbK69eLyqhWc3kNwADArGV1DK31xvfsrW+qIDH4x/fvb1Vt/ckQaFOu/XAxrXtew04kmsQE7iFYsDqraySVFNLUv0K4LmZwZsW59kiy/F8U7hUYHB5FQLSQTE0AgCweehrJsfDJkcGeNWQyWSHDx/+4YcfcnJyWrduPXz48EfoRKfTPZSpKAI4UWCyeWmaRbW1MoPLeyTPSLNIJeSp7sofxzCgWXQs35ikUxfY3ddNDgGBP5F3g3KVTK94YVMgwHpGlBNE4FY3r1/35ZdfTV2+nsvMPbpj84R+3cU8ouj4PrJy3NEdW7jG5Tw+7wcPxxLCbk98CAicW62llckFZIRMBAAYBolhSp1MqHtJdUcCvNA4nc6+ffuuX7+eJMmPPvpo9uzZOI6/3bNfusl9Lf38iT07AGDWitVnzd5gkWDFnGntBn36fwOHAcCGhb/U0Mo6jR2zZNWaQlw849NBbqcjsmr1tu/1YSjq6w97Xjl3xm42vTvsMxpHXCrkdZMjx+riSsQFBB4sEdQPVWAAMQoxADgppqpayiJTsdNLs6iySsI1ptiXMBU3wBMnJibmv//977M8IgaQGKY8nGe8YrC7KKbA7qZZVEsruzu6AoCqKgkguFBqPZxrQAACEk+O1Mj5TyAeGDhw4LJlyy5evAgAISEh48ePf/w+X1wCAda/yYwZM1RBwTJlWQKjJihIrlACwPAhgwd88MHFk0dJkuzXr5/Pdu2R4aIrDgyDQHQV4Pnk22+/Xbt2LQB4vd5ffvklPj5eJBJ5omrKNMFVX6u/9NAZtQC38mUIAeCgUcq9Hg9PIEAIFWRfb9egQ1FR0Ud9elJU2UxHSda1eV+Munr+TN71TADYsGDOxC/GBwkJD82kmSmKYbnoSsQjmkdqJH5ZiVyMhQCsXoZmEZd3FaMQpeYarxrsOqnwASXlAQL8WyiFvORIdWqukVNlq6WVcX5r96SqWlJod+tdXgCorZU9kegKAFQq1enTp3fv3u1yuVJSUh7sL/LSE8jB+jeRaEMm/W+NVKHcvHh+4c2s6glJpwtNCIFAIFixfLnVarVarQsXLrzbQDpAgJcSX247x2+//TZo6DCXqyxfipQprHwZAJA4Xl0lad53iEQuP7xtE0PTQ7+ekdSuU3Z2ti+6AgCj0VgrXMtFVwAwYcKEWqFqAY7xcdy/ZClELJDcq+YDA0jSqeqFKLisdoWAlxypjg9TBqKrAM8tCIGvbM032up2u69fv+7xePxb/l1i5fQRuWWDy/ukzkEoFHbs2LF79+6veHQFgQDr8bl27drKlStPnTpVbvvx48c/++yziRMn3rx58377fvL1dIlc/ufcGUt+mLJ1wc8CAs+zuUucHgDQu7wujPe0xegDBHiuqFWrlv9qenq61Wj4snfX4tzbNxGGQYso9alCC4OgpkY6+O1WdeQ8HMfPlVir1XktNjaWa9a2e+93e/WZM2fOX3/9NW/evN1n0/8z7gsA8DDorMlTlneFYQCQbXH+XWK95/lI+aS/2K9CwIuSB27JAPeGubMkn3lWFfoIoXXr1n355Zdrtu1MzTVQDFtJKeYReKbRcUlv27RpU3h4eJUqVcLCwrZu3crt8neJ9ZrJISDxltHaOkEymkVH8ozGJ5HkHsCfwNDIY7Fo0aKBAwcyDAMAI0aMmDVrFrd9+/btHTp0YFkWAObNm3f27NmoqKi7d3+/WfyqHXuw0rzRY8c1G/AJxaIqKkmIRKB3eY/mGQHA58rMMExRUVFwcPATVKgPEOD5wW63/9///d/evXv9N3L6CGZ9yeHtm7oOGn5rI6QVWRJCFQY3VUMjBa0MANR2j5Nm1FLxvn37JkyYQAul3UdPEhKYg2JaisXV9ZaCnXtOaEJbRGnSzW4nzQKAmEfUC1acLDLTDHvN5BDziCqBso8Aj0q2xXnFYG92a675qtGRa3UlR6r5T1/eafDgwQsWLMAJYubm/REs4mYGYxTi1DzjFYN90rc/mEwmADCZTL17987OzgYAuYAUknizCI2cT8rVUgzDuDz3p32qrxrPkQ5WTk7O/Pnzt2zZcuPGjbCwsPbt23/11Vcazb8vOH5PHSwAYBhGq9WazWbflszMTM6qtl27djt37vRt/+abb3y5fqVOr5NiohVl78EFNjeGQZhUmGN1nSk0YxhU18gyjXaaRXWCZNXUUgA4evRo9+7d8/LyCIKoVq3ajBkz2rVr538mnA7W8zweG9DBekV4ZB2sSZMm3W2H0Lhx46ybOVNWbtGGhpX7E5/AqxKe44f+ioiIaNa8xblSW1yQ3KfuhgDSiizZFqeIJOotnB32wzfOqBjv1atKAa/IaL5h8zI4GR+qEPMIs5u6bna4aTYxTPkMnoUVIaCD9aLgr4N1qtCce0vsMMfquqS3kTiWHKm5Z475E0Sv14eEhHAv82HRsfEtWv/63WTul9bspq7lFSZWjfZvn5mZKZVKQ0NDaRb5l16VW32uCOhgPQH69OmzcePGH3/8saSkZOXKlfv372/SpInT6fy3z+u+MAzjE9vlsFrLJhruV6pKseh4gelskTnb4gSAApv7ZKHpyM3STl27zZ7yZW2lAAFk6G1c9QcXXSGEevTowZkbMAyTkZHRqVOn9PT0Z/ABAwR4Zly/fr3cFrFYPGPGjNTTZ7WhoQDAUNS0YR8wdJmkgpdhjxaap8379aNhI+Zu+yvf5s4w2BwU46FZAMAAGoQqKinFLpoxuSkAEJK4UsDLt7lJDKul4CdHqrloTCnkxYcqm0Y8i5GGAC8xCWHKSLnIQTH7b+q56KpphPppR1cAQBAEfssZs/Bm1vbli3w5u0ohL75yZP369W+fZEJCZGQkt3zPwvYAT5bn6DelTZs2p06dSklJkUgkiYmJ8+fPv3r1KldS9HzC5/P9NdaTk5N9X+Uvv/zS9xJfv379fv36ccs8HEsMU2IYllZkSSu2nCw0MQw7Z/zIzevXzZgxY9IX4wgMAwAMQC4ou0nsdntOTo7/cSmK2r1791P+cAECPFO6du3qW8YwbMiQIenp6U2aNIkOCZLwSIQQTpKDp0wjSNLjdNn0xad2b1VogiYt/vP7Ndtja9a5ePJo8cVzh3IMh3INXs7ULCNDfPyY9vQJcUEeAGBuT9HOvdc378zadQAxAZXdAE8YzpFMKeBxX78GoQqN6FnYdatUqtGjR/tWR44cyY30MAxz+vTpS5cubdu2bcCAAfHx8QMGDNi6dSsWEMJ9hjxHOVhffPGF/2qlSpUA4AEZ4o+G0+mcNm3aqVOnatSoMW7cuMf09Jg1a1Z8fPzx48erVas2aNAg33c3Li7uypUru3fvlkqlbdq08Z9eDJUIknSq4/mmLLMTA5g7fuShzesAoHr9hLc+/pxmUYRMmG9znywwNwpXhkmFMpmsRYsWBw8e9D9unTp1Hue0AwR43ujcufMff/zx22+/4Tg+ePBgX7zFJ/CW0dqBn4/rNPhTuVpz/eLfM0YMcljNIgHfYDC82aMvEEB5vedSD1Yb+imPZqIVIh6BF9rd4lGfV9u5vdqt/rGC/NB2bUIBAMCSnQ3PQe5BgJeMTKPd7CnLE79YauN8Kp/Bcb/55puUlJTz58/XqVPn9ddfB4CCgoI2bdpcunQJAFJSUjZu3CgUlknzeL1PrFowwD/yHAVY5eBymLgw6wnSq1cvzvZy+/bt+/fvP3PmjG989RHAcbxv3759+/a9+09CobCoqOjq1atGo7FXr17+R0FQVkmLAIQiMQCIJNKx8/4nkkhrqES1gpVcPtbJAnNKpSAhSaxZs+aTTz7ZsmWL0+nk8/mjR49u27btI59zgADPJz169OjRo8fd20tKSxv/P3v3HR9FmT4A/Jmd7SXb0isQqghBOCC0ICiCDWwnpyBiuRP1LCg/70A5ETjRs4AntrOAgMp5NrCAoFISiqGEEjoGSE+2993ZmXl/f0xc1xBCyia7G57vH/nszM7sPjvZd/fZmfd93utvBgBCSI/LBgweO37DRyv+cNXVuZcNEDaQSKXTnpwLACkq6eBUXa3b/3O1fXiQwynNUac5ZfWUmlxiETUiw3DO4S13+gorrGOyDJ2TYxUUFBQUFIQWn3vuOSG7AoDvv//+gw8+eOihhzohDNRIjCZYZrN5/vz5WVlZoSliOI5jf+1+IRCLxa3NjVwuV2hScQA4cODAwYMHBw0a1Pxe5FctfyJCyKRJkwoLCwHgrbfe2rNnz7///W/hrhq3v7jGQQjJkIvK3cz98//p9/m2fvXpO//4v/97+h/9eg8nhGRp5IRoCSEyWkQISUxM/OSTTwCgvr5erVYrlcpGwbQhwk4WOxHiGfLI2r59+yuvvGKz2W677ba//vWvEX98H8vttwZSsnJ2bvx609rVf39zxX3PLJbK5XfOekSs0Z09crDq7JlR1zfMBGryMMfMrpNWDyGk8rU3VUo6QSp2vvxqwrJXvRlZuz5dPyLDwPm8JCEh4nGiS5wnyAr9rowKaaJSyhMweQNBnkSlqofQZ/dCi6jTxGKCFQwGp06dajKZNm3aFCoEtWTJkkWLFoVvNn/+/CZPHTWDZVm9Xi+MWQUAoTPgRd98ZrPZ4/FIpa24oF5aWipkV4K333579uzZEomEA+qQjyYESjete/CfC7L6Dfy/19/7y4IlB4q22ivO5HXPDAUjAmAIVemyNnpkq9VqtTZe6Xa7eZ53u90Qq1iWtVgsfAxMMxLq44na79ixY9dcc41QwLCwsJBhmIj/UA6wPCWW7tm8YdmTD3Mc+8LD98x7a9VNd90j1uiSlNKrr796Z5XVG2zoU8UDHLe4AeCKFG13XRoAlDt9biK+DEAkFjuMKds5enCKVko1JPp1noAnyIUqXVU4fQCQhZWuUOsNStH2MqiFGg0UwNB0nS/IKZsqYNsJ7rrrru+++064LZVKb7/99qiEgaKTYO3evXvEiBGhxfATG4SQu+++e8uWLWvWrLnyyitD65955plnnnmm/U/9/vvvT58+3ev1isXiV199NXyExYVIpdImyzQ0o1HNXLVa3a1bN+F8m8bHvPPByn88+iAA1NVtfumR+3mes5vrJ98+tdQv7mtU9zGqAaDK5T9cY8tL1nYPq3N4IXFRpkEmk2GZhi7m66+/Dn+r/+9//4t4gqWTS8Z3S0wfPnDHyBFlZWV9M9Pyk5Xd8oaWO31JCum2CotQ9MQol1a4fSYPAwA0RamkNADY/MF9Nfa+QAGAXCzKTlCUO30HbP6hBhkABHmyp8bOcDwhJFevqnD69tbYKYpKVHZS1xnUxYTPB0ABRCu7AoA//elPEolkzZo1CoXir3/960Wv0qAOEp0EKz8//0JXi5544olPPvlk+fLld955Z0c89c0331xZWXn06NHc3NzU1NSOeAoA6Nmz52OPPfbaa68BAE3Tr7zySuhqplEhXbHspdCWe7duBgC9Xn//vffaKDhidhEAjVRcXGMDAjQOnUUxTKj6dqHF5jmdzjfeeKOsrGzUqFEzZsxo8nK/N8gpJbRaKv7DH4Zs377dG3ZKQKionp2g8LHc4FRdnSdg8QYBQCMTuwLsrirbyAxDklLay6BOnDkNxo0AtXpImo4WUQaaFxqVREQNT9fvqrIerHfa/MEKpw8o+EOaDrMr1AXceuut4SNzUVTEUKFRAFiyZMm8efMWL17caERhdF2o0GiIy+WqqKjo0aNHaKSGYFtJaU1VxZA+PXv16gUA5xzeBJlEL5dMmTJl/fr1oc1eeOGFv/zlL3q93uJjdlRaWZ4IXwBD03VpKrmTYS9aTCUuzmBhodGuhxAya9as9957j+f5vLy8b7/9Vq/Xt6TQKMdxw4cP37dvn7D42GOPDRkyZPXq1Uql8pFHHrnqqqsA4KTVc8ziEvIkADhY7yx3+EZnGbZ89/XmzZtTUlIeeuih5ORkAsCw/MYz9TxPrkjVdtMqy+zeA3UOCS2a1CNZct5PlEaNxeRldlZZOZ5QAEPT9ZkxMA86FhqNF+GFRuMFwzBWq7XjTi50hPgtNBpDfbBWrFgxb968OXPmxFR2dVHvv//+I4884vP5EhMT//vf/44fP15Y7w1yNlWioneiNl0PAGcd3v21Dhktmtgj+fXXX6+qqtq3b59cLn/mmWf+9re/CbsYFVKFhHYFWAKQlaBIU8l3Vlmt/uDoTEPnlFRBqFUoinrnnXcWLlxot9t79+5NUVQLKwP//PPPoewKAN54443QEJZvv/12586dQ4cOlYgojic7q6wjMvQ17oAwlceaVasefeB+YcvVq1cfPHhQrVbLxKIhqVqOJzlaJQAIfaoUYtH52dX5GI7nCQEAAuANshfdHqGoC/KkuNp2WaJG+O0dYPniGtugFK1GGkNf6AhiqtDoK6+8AgAvv/wyFaYjxiVFUF1d3YMPPujz+QDAbDbfddddoTOCSgk9MDmBI2RXle1gvXN/rUNEUUPSdGIRlZ2dvXfv3qqqKovFEp5NVrn8why0FECF0/fjWbPJy2ik4gRsNiiGpaSk9OnTp1XDMxsN1AgfIMyy7Oeffw4A3XXKvOQEjic7Kqy/2DwysWhMtnHpC/8MbVlWVhbqyZupUQjZlaCHTpmmvvi5qEqXv7jGBgB9jGpaRJWaXGX22J06Al0KeAKH651+tmE8EMeTg/XO4O+HB1U6fXWewI5Kq90f9LP89gqLycucsnqaejwUTTGUYJWWlpLzLF++PNpxNae8vDwY/G0G8urq6vBf8D10ysGpWp6QX2weEUUNT9elqn67zpienq5U/vaVUO3276mxAYGeepVwjdAdZBVienSmQYKTeKAuhBASXnv6fN27dxduHNzyvb3qnPCT5TKjJkEqbnQdXK/XtzkMlicH6xwAMDRN3z9RMzLDQIuowyann8U67/GnvLx83rx5AwYMUKlUPXv2fPTRRy0WS7SDaotf7J5TNk9hhcXPcsIZ3F9snkP1rvBtuuuUfY1qhuMLK63bKywuhk1TywelYPGRmIPf3O0yaNAgoX+V4LrrrlOpVOEbhP+ob/4nvkpCS2jRkDRd/ySNVtrQ6SpdI8cp0lAX43K5jh07dqF7b7755nvuuQcAFixY8M3eUl1GjtftAoDDJqfZy7z00kuhzkm33nqr0FurbYSqRcPSGvpdJSmlIzMMIzL0cuzkHofibirbC8nVqdI1chfDbq+w7qyymryMTi4ZkKRptNlliZpeelWQ490Mm6yUDU/XibDCX+zBL+92kUgkP/zww8yZM0eMGPH4448L5UBDhH5XFEV11yl5Qn6uspm8F5ymQCuTTOyenKGR76q0OQJBtVRMAZyxe6rd/o5/HQh1noSEhLy8vEYrpVLphg0bysrKvvjiC6HmXC2R3XD3n+0W09w/3bhiybMsT3ZWWQeNHHP69OnPP/98+87dr32wOjT20BPk6jyBxs90MTq5JCOsV3uSUpqsbEU1FhQ7Ynkq27MOL8f/NpjsjN3LX3homYiCYWn6DI3czbAmL6OTSUY3NRN5gOPrvA1veJs/6Ahg98FYhAlWe2VnZ69YsWLnzp1Lly5N+LVC9JYtW/4+/9n9NTYAyM/QX5GiHZSi5QjZU2PjLjxsUyyifq6y1XsDOrlkXLZxSJqOECiutpt9OHvUJaRVFzt++OEH6vfaOb1mJ+B5fvHixfn5+RqNJjEx0WAwDBw48LPPPps0aVLo4iAA1Jw5Za2ve27m1MrTJ7/58F19wCGlRXIxnZKScvMttzBpucU19nKnDwCWvvHWf38+vP1s/bK3342pYdGo0zz99NPhVw86aCrbNjhj9+6vdeyssrI8AYCSOkdJneNAnaOZXQiQINfwNmYJ4c97SwdYvrDc4gywaWp5b4M6yPM7Kq2OQPC8R/qNj+V8Yde+3QzbMCc66kjYezry1qxZM2PGDELIqP0H1EplwdvLQSXroVNSAFqZmKYoAGA43s/xod7rfpbjCKgkdK5exfBkZIZeQouyExQiijpj9+hkFynTgLqSGTNm1NfXL126dPTo0UePHr3nnnu+//77kpKS8B57jRw+fDhe5v82m83XXHNNSUkJAIwfP/7rr7++0Ov607UT/nz9WI/LCQDTp08fN7Avy5PqygqbSJSZmXlZovrnavu+GvvOPXtV/YfqklK2fvW/5fNm6+SSmTNnduYrQjGog6aybYPMBPlZh1eoBqKWiM86vHKxqJdBdaHthXFRws9shZiucfsLK6wFWUaZ+LezIbWegJNh09Ry4cogRcEJi/ucwzcwuelvigDHF1ZYCSFjsoxiAB8P+yssMpouyDa2ZKQtajM8gxV5r732mvAzesd367//bK0wJAoAuuuUBoUUAFieFFVahZ8gAOBnue0V1sIKi4/lUlSyK7ONoRPCmRr5mCyjGNvApSSWL3a03/PPPy9kVwDw008/vfXWWxfa8o477jiwf9/KlSu3bdu2atUqt9t93aSJOTk5WVlZU6ZM0YthRIZeRFHSzJ6JaRlF36574+knCM//9NNPnfVSUIw6fyrbKJKIRKOzjAaFxOxlzjq8MrFodJaxmXoKZ+zeek9AJ5eMzjQMT9enq+Uuhj1q+V0n9xytYkSGPtTvqn+iJj9dPyD5gp3cpbRIL5d4glxhhcXkCx7xED/LGxQSzK46Gp7BigDy+87smt9PJavRNO6fKBZRSUrpKatne4VlaJruYL3TzbDpGrmMxt61CBrVgYudix0RUVFR0cxiIz179gxVh3/ppZc2b94s3F6/fv3y5csfemy2lBYJFz4O7tjKcxwADBkypEPiRnGiyals33jjjTfeeCN8s4cffrj96RfLsmazuclJCBqR/Hp1Tkp4p9nkuXBiowTIlkGKJGiprwOALApoGSRz3pqaxh3268KSLgqgttmpaDMA/BIwBbmfa50AkCyBFL6Jx4xNgUDA5XJxXPSH96alpbVqe0ywIuBAnYOmqIHJCQDAE3hy6TvpK977aNmLAFBQUNBkMx6QlEBT1HGLe0elFQCS5PSwND3+nEDna8nFjnHjxtnt9sTExEmTJi1evDiWS0tPmzYtdDZOLBb/6U9/auGOp06dCuPzQEoAACAASURBVF+srDNtr7D4WC5JIa1z+x5c9DIh5HDhT8uWLauqqlq8eHGrZmdHcaQNU9nefvvt48aNC3+QpKSk9k99wbIsz/MX7fV42OyuY/xysUhGixwBtoyVDEtJaGYatKTfL0aqfrkiyG2rtPOE0BQMTDMoxHFz/crv90skktjvXXq+2JoqJ+LcDOtmuFR1w8gguz/I8iRR2bpP3uanyvEGuS3nzAGO76lXXZ6kKa62V7v9Cpqy7tmaZDSMHz9eLG46i6232jefNSlUap7jXnpwxup33+7du3erAgvBqXK6KrPZnJeXR9P0iRMnQj/HwxUWFm7atGn69Onp6em7d+9+6KGH/H7//v37k5KSzt+4E3i93otOlbNu3bqVK1dKJJIHH3ww/GvPZrOtXLnSZDJNmjSpoKCg0V6ff/75bbfdJtymxeLP9p/kpfLsBMWQNF2t27+7ysZx3N/+eF3Z0cMAMG/evH/+85/QlNhvLDhVTpvNnj172bJly5cvf/jhhzvh6VoyVY4wcZNcLBqTZZSLRUUVVps/2E2rHJyq7YQIQ9wMW1hh9bGcUizysrxKQhdkG+Nl2s34nSqnKydYHE82nzX5WX5Yui5dLbf7g4WVVkLIVd2SVK2Z51xIsEQi0dy5c9euXatQKJ588slZs2aFNnAG2MIKS4DjVRLaE+TUUnFBlqH5ajp+ll938JREnVBbfjY1u5vTatn72arlr/yrba809r8zMMFqg2AwOGnSJCGFCv853ozDhw/n5eXNnTs3lF4sWrRo0aJF4dvMnz+/47qB+3w+n89nMBhau6PH47nhhht++eUXAKAo6uWXX/7jH//YaJsvvvjiv//9r0gkuuuuu8ZMvN7CUTnShpmb1377/aFTZ9a8+rywZV5e3tdff93kE7ndbp7nExJityqj3W6XSCSNKupFV1ZWVrRDuLjOn8q2JQkWR0hJraOPUS30uwryfEmtc0CypjOTGz/Lbzln9rFcd53yMp2iuNpqYohaKh6XkxgX3bAwwYpR5U7fvho7RUG/RM1JqyfI8X2M6v6JjTtFNU9IsF5++eVnnnkmtHLjxo0TJ04MLToCwZ/OmQkBiYia0D2p+eyK5cmWc2YXw+7e/N2rsx+c/uTcyffM8rtdN+fltq3VYYIVp5q/2DFt2rS1a9euWbPmzjvvbPlj5uTk9OrV64cffhAWOY5r1HeBpumW9Bppm5acwWrSxx9/PH369NBi//79Dx8+3Ggbi8Xi9/ub/D77/vvvr7322tDin//853feeafJJ4r9xhKDZ7BaNQ9SVKxYseLee++dM2fOSy+91GlPGkeTPR+qd7I8uSJVG2QYi9VaReRyMX35efVLY1P8JlhxcxW2bbITFEPT9QTgiMkV5PheBlVrs6uQrVu3hi9u27YtdJsAnLR6hC/HIE9OXmxOKLGI6qZVakXce8/+jWODH7648Mt3l2uAxRLSl5r8/PzwiaHC73riiSc++eST119/vVXZFQBQ1O9+NdE0Lf09mqap2GMymcJfhUKhCL+XEHL//fenpKRkZWUVFBRYrdZGu0+aNGnhwoVKpZKiqGuvvfbFF1/svNAvAe15k3eOeJzKtjMNTE4YnKoV/pEUwJA0XbxkV3Gt63dy10hpMSUKEp4CMCra3u916NChobMCwqJwgwDsrbFXOH1qqXhQSsLeGvtpm4cCaGbQLAD0Mqh6GlTFP+9+7733/H7/1FtvGDHksjbHhrqYJUuWLFu2bPHixa3tSlJaWlpeXt7anCwWNJoF4b777gtf/Oijjz744APhdlFR0T/+8Y9Gg8IAYP78+X//+9+DwWAzBcNQV1VaWhrtEOJJHKTMXUIXP4PlCAQLK6xBnk9TywGguNpe09aZZ+bPnz99+nSxWKzRaBYsWHDzzTcL6/0sZ/ExQr+rZKVsdKZRRovMPoZrZjYEAACgAHr27PnCCy8sW7Ys/DoRusStWLFi3rx5c+bMabIrycaNGymKWrNmjbB4//33r1mz5ty5c263+8cff7z11lvT09Mff/zxzg05AhoNEkxJSQlfPHnyZPjiiRMnmnwQiUSC2RVCKEZ05QSL48mOSivD8X2M6hEZ+iFpOkLIz9V2b7At5TQUCsXq1at9Pp/T6Xz22Wd/Wy+mC7KMoV7tCTJxQbZxdJahmVG4CDWjVRc75s6dW1RUNH78eIPBMHPmzIKCguLi4s7vrEAI2bt379atW/3+VvyAKTW5hLluAOCWW2994LkX/3DlBAAwGo2NRhHeeOONdFiVuJtuuikSUSOEUAfqypcIaRE1OFVn9TGXJWoAIDtBAQAMxytbM4SwkSZrLjR6wGaq9CJ0Uc1f7Jg0aVJ4F6vc3Ny3336744NqTiAQuOGGG4QL6JmZmZ9//nlLOrk7A+xpm0foe5atVTyw4MUKd2DYuAnf/Vv7j/nzjUZj+MbDhg1bt27d0qVLPR7PHXfc0TmD8BFCqD26+CjCRggh77///pdffpmQkPDEE0+E+lE1r/k6WLEg9gdG4SjCLuz999+///77Q4s333zzF1980ZIda9z+n6vthBCjQmr2MQoxPSbLoO7g3yex31hicBQhalIcjSIMYRjGarWmpqZGO5BWiN9RhJfWuZZ///vfoe4p69evP3jwYGgiDoRQSwSDwY0bN5pMpnHjxnXv3h0A6urqwjeor69v6UNZ65WWOrch0+xjZLSoE7IrhBDqNF25D9b5wkcqeb3er7766kJbljt9AZb/bdEVCF6s0zpCXV4wGJwwYcLkyZPvu+++fv36fffddwAwderU8Ak3w8tZNWPRokXdu3f/bGPDyFyG463+IAGodP3Wi4snUOVq46gUhBCKrksrwcrOzg5fzMnJaXKzWndgb419e4XFz/IAUO4nhy2e4mpbZ4SIUAz79ttvQxXgAoHAP/7xDwDIzc3duXPnrFmzpk+f/umnn4YmtGlGWVnZggUL/rLgxUl3zrTU1vzv3/+iKGpfjX13pa242naw3gkAPIHiGtvP1bYye3xMSYsQQuEurRPyL7744sGDB4Uh39OnT7/QbOrJKmm6Wl7t9hdWWFJUsgo/kYhEl7W1QilCXYbP52ty8fLLL3/rrbfg10ruF32c2tra9O65BTfebKmt+ceMW83VlS89v3hvncsVDErFol9sHgDwsVy1y6+WitPUsdv9ESGELuTSSrC6d+9+9OjRw4cPa7VaoftIk0QUNSxdX1xjq3b5XQwrpmBYilovl3RmqAjFoMmTJ/ft2/f48eMAQFHUnDlzLrqLm2HPOnz9kzRC2RK7P1jjCQwbNkwrly66/06bqb62/OzMmTMztSqJWKyVSYI8X1hhFXIstVQ8JssQL1PSIoRQuEsrwQIAmqYHDRp00c1EFGil4moAAJCIqJZ/xNfW1q5duzbAMH+87bYePXoIKwlWzkVdgkql2rVr14oVK0wm06RJkxpVq2rS3lq71RdkOP6KVK3DHyystAY53iCXbNmy5cUXXywrKxvx14cee+wxAEhRyQBARkQ6ubjWzQFAokKK2RVCKE5dWn2wWu6YxX3M4paIRMlKmY8jP9c6Axx/0b3OnTs3cODAnw4d9yZm/2HYsH379gHAEbOruNp2KVXDQF2ZTqebPXv2888/35LsCgCGpumVEvqsw/vhj7u+OXImyPHm4weTldKkpKT58+fffvvtmZmZoUuNhMCeGlutO6CWimVi0VmH91C9syNfDUIx5Ouvvx4xYkTfvn3//ve/MwwT7XBQe11yZ7BaosbtP2Z2SUSi0VkGrUy8/UydNcjtqbGPzjQ0v+N//vMfu9M5dvJt3fv1f+Rfy/+9fPmcf712yuqR0iIvy6naUeAUoTilktBjsowbjp1LyOwGAF/85/WPXl0i//DDcePGjRw5srKyEgAyMjJ27NiRk5Nz2OSsdPmFK4NBnhRWWE7bPEoJ3VOvivLLQKiDHTp06JZbbmFZFgBefPFFAHjhhReiHRRqFzyD1YRUtTxXrxqdZdDLJSKK6quistSyvGYnbxbQNB0MBJ675/YzR0vzRo295qGnhOxqTJYBsyt0yWJ5nv21yok+KZkSibZt2/baa68J2RUAVFVVLVu2DAB66lUpKpnQ7ypBKh6TZUxRyXISFFELHaE2OWl1Vzh/GxFyxOSq9QSa3+XHH38UsivBxo0bOyo41FkwwWoCBZCXnBDq1U4BDExUtWQCnFmzZmVmZrrstgX33O60WhQJOgAYmWHQyrCDPLpECROuSxXKjR+vNNdUj7t56qzn/jV06NBGgw2FRaWEHpX5W6/2BKl4VKZBQuPHFIonboY9anbtrbELU20eNjlPWN0ltQ6u2Z4iAwcObGYRxSP85Iqk9PT0gwcPvvvuu2/+98sEQ8NkakctrubbFUJd2IE6J8PxX767/N2F8569+zZLbc1Vt91x810zH3zwQZWq4cKfSqV66KGHohsnQpGiloqHpeuF0m47K63CdYyRmXqaam6w01VXXbVw4UKFQgEA48aNEyZ9R3Ht0pqLsG1aOxfhUbPruMUtoUXD03VHTC6bP5iikuVnXKR1tUfsT6+GcxFeIoQ6WOGTPVud7vuf/NuX770pLGb37L1h+47L0hIB4NSpU2vWrCGETJ8+vXfv3p0TYew3FpyLMF40Pxdhjdu/u8pGACQiUUF2S69jBAIBv9+v1WojGulvcC7CzoSd3COMJ8TsZYR+V1qZRJcpKaywuhiW4XgccI4uQYzXHcquAKD89EmF3wWQCAC9evV67rnnohcaQh3I5g8KZy9YnncE2BYmWDKZrOU/5lGMw0uEESaiqJGZhrHZRqE5CZnWmCwjZlfo0pSamnr99deHFsePH99MjV+EuoZSk+u4xS2lRZcnaSgK9tXYw/u8o0sEnsGKPLGICu8RL6VFUkyu0CXsf//736uvvnrw4MHLL7/8iSeeiHY4CHUsN8P+YveErmOoJeLiGttRsytDIxd1WEcRFIMwwUIIdSyFQvH0009HOwqEOolaKh6ZYZDSlHAdI10jz6cMGpkYs6tLzaV1ibC0tPS+++6bMmXKihUrsHc/QiFOp3PevHk33HDDvHnznE4sno5QuyQppeGdrlLVshgphXj8+PE5c+ZMnjz53XffxS/BjnYJncGqrKwcPXq0w+EAgPXr19tsNrxagZBg8uTJ27ZtA4Bvv/12165dW7ZsiXZECKEIq62tHTdunNVqBYCvv/7abDbPnTs32kF1ZZfQGax169YJ2ZVg1apVUQwGodhx7tw5IbsSbN26tby8PIrxIIQ6wjfffCNkVwL8Euxol1CClZmZ2cwiQpcsg8GgVCpDi0ql0mC4yLSbCKG4g1+CnewSSrBuvPHGadOmCbezsrJeeuml6MaDUIzQaDRvvvmmVCoFAKlU+uabb2KVS4S6nokTJ959993C7YyMjFdffTW68XR5l1wl91OnTplMpiFDhrS8mFtrK7l3vtgvTo2V3GOfyWQ6ceJEnz59kpKS2vwg51dyjzWx31iwknu8aL6Se2xiGGbfvn08zw8ZMkQul0c7nBbBSu5xo1evXr169Yp2FAjFnKSkpPakVgihuNC9e/f4mionfl1ClwgRQgghhDoHJlgIIYQQQhGGCRZCCCGEUIRhgoUQQgghFGGYYF3cd999V1dXF+0omlNaWrpnz55oR9Ech8Px+eefRzsK1OHOnj37008/RTuK5hw6dGjv3r3RjqI5RUVFp06dinYU6OI8Hs+nn34a7Shax2KxrF+/PtpRtE5VVdWmTZuiHUVbYIJ1cR988MG5c+eiHUVzduzYEePvP4vF8vrrr0c7CtThjh49GuNfOUVFRT/88EO0o2jOunXr9u/fH+0o0MU5HI6lS5dGO4rWqa6ufuedd6IdReucPn169erV0Y6iLTDBQgghhBCKMEywEEIIIYQiDBMshBBCCKEIwwQLIYQQQijC6AULFkQ7hlhHUdQVV1yh1WqjHcgFURSVnZ2dm5sb7UCao1Qqhw8fHu0oUIdLSkrq379/tKNoTnZ2do8ePaIdxQVRFNW3b9+0tLRoB4IuTi6XjxgxItpRtAJFURqNZsiQIdEOpBUoijIYDAMHDox2IK12yU32jBBCCCHU0fASIUIIIYRQhGGChRBCCCEUYZhgIYQQQghFGCZYCCGEEEIRhglWi5SXl8+bN2/AgAEqlapnz56PPvqoxWKJdlAAAPX19XfddZfBYFCr1RMnTjxy5Ei0I2osZg8daqfW/md/+OEH6vcSExM7KLZWtYvOb0StOnSdedxQa8XX51vsf180El+HtwkEtcDYsWP79eu3ceNGt9tdXFzcv3//3r17ezye6EbFMExeXl5+fv6ZM2fMZvMdd9yRmJhYXV0d3agaic1Dh9qvtf/ZzZs3A8Dhw4c7OrBWtYuoNKJWHbpOO26oDeLo8y0uvi8aiaPD2yRMsFpk8eLFbrc7tFhYWAgAH374YRRDIoSsXLkSAI4ePSosulyuhISE2bNnRzeqRmLz0KH2a+1/ttMShVa1i6g0olYdOkywYlkcfb7FxfdFI3F0eJuElwhb5Omnn1apVKFFoUrhuXPnohcRAMD69etzc3P79esnLKrV6vHjx69bty66UTUSm4cOtV/M/mdb1S6i0ohi9tCh1oqjf2VcfF80EkeHt0mYYLXFxo0b4dd/dhQdOXKkd+/e4Wv69Olz5swZn88XrZAuKkYOHYq4Fv5nx40bJ5FI0tLS7rnnnqqqqo6IpFXtIhYaUUsOXSccN9R+sfz5Fgtv9XaK5cPbJEywWs1sNs+fPz8rK+uWW26JbiRWq7XRBD46nY4QYrfboxVS82Ln0KHIasl/ViaTPfPMM0VFRVarddWqVUVFRfn5+SaTKeLBtKpdRL0RXfTQddpxQ+0U459vUX+rt1OMH94miaMdQJwJBoNTp041mUybNm1SKBTRDiee4KHrqlr4nx0zZsyYMWOE2xMmTPjiiy/y8vKWLVv2z3/+s7MijTktOXR43OICfr51qDg9vHgGq7Hdu3eHj4gOv4sQcvfdd2/ZsmXlypVXXnlllAL8jV6vdzgc4WvsdjtFUTqdLlohXUisHTrUKh3RKAYMGJCVlfXzzz9HONZWtosoNqK2HbqOO26oeXH01dCkOPq+aCQuDm/TotW7Ph49/vjjALB8+fJoB9Lg5ptvzs3NDV9z00039ejRI1rxNCPWDh2KlPb8Z3NycsaPHx/xkFrVLqLYiNp86DrouKE2i4vPtzj6vmgkLg5vkzDBaqnnn38eABYvXhztQH6zYsUKOG/Y7eOPPx7dqM4Xg4cORUR7/rOHDx+mKGru3LkRj6pV7SJajajNh67jjhtqm3j5fIuX74tG4uXwNgkTrBb54IMPAGDOnDnRDuR3GIYZMGBAeOE4o9FYWVkZ7bh+JzYPHWq/5v+zGzZsAIDVq1eH1tx3332rV68+e/asy+X64YcfevfunZGRUVdXF/HAmm8XjQKLSiNq1aHrtOOG2iCOPt/i4vuikTg6vE3CBKtF+vfvf/7V1YcffjjacZHa2tpp06bpdDqlUjlhwoQYrEYYs4cOtVPz/9nzE6zTp08/8MADPXr0kEgkmZmZ999/f1VVVQfF1ky7OD+wzm9ErTp0nXncUGvF1+db7H9fNBJfh/d8FCHk/BeAEEIIIYTaDEcRIoQQQghFGCZYCCGEEEIRhgkWQgghhFCEYYKFEEIIIRRhmGAhhBBCCEUYJlgIIYQQQhGGCRZCCCGEUIRhgoUQQgghFGGYYCGEEEIIRRgmWAghhBBCEYYJFkIIIYRQhGGC1XV4PJ6+ffsOGzYsGAwKazZt2iQSid54443oBoZQzMJWgxBgQ+gYONlzl1JSUpKfnz979uwXXnihrq4uLy9v+PDh69ati3ZcCMUubDUIATaEDkAvWLAg2jGgiElLS1Or1QsWLBg1atRTTz1lMpk2bNigVCqjHRdCsQtbDUKADaED4BmsroYQcv311//0008Mw2zevPmqq66KdkQIxTpsNQgBNoRIwz5YXQ1FUXfddVcgEMjLy8PmgVBLYKtBCLAhRBpeIuxqamtrp0yZ0rdv3wMHDuh0uvz8/GhHhFCsw1aDEGBDiDRMsLoUQshtt91ms9l27drFMMySJUsmT56ckpIS7bgQil3YahACbAgdAPtgdSmvvPLKU0899dNPP40dO5ZhmPz8/EAgsHfvXoVCEe3QEIpR2GoQAmwIHYGgrmLfvn1SqfTpp58OrTl+/LhSqZw1a1YUo0IolmGrQYhgQ+gYeAYLIYQQQijCcBQhQgghhFCEYYKFEEIIIRRhmGAhhBBCCEUYJlgIIYQQQhGGCRZCCCGEUIRhgoUQQgghFGGYYCGEEEIIRRgmWAghhBBCEYYJFkIIIYRQhGGChRBCCCEUYZhgIYQQQghFGCZYCCGEEEIRhgkWQgghhFCEYYKFEEIIIRRhmGAhhBBCCEUYJlgIIYQQQhGGCRZCCCGEUIRhgoUQQgghFGGYYCGEEEIIRRgmWAghhBBCEYYJFkIIIYRQhGGChRBCCCEUYZhgIYQQQghFGCZYCCGEEEIRhgkWQgghhFCEYYKFEEIIIRRhmGAhhBBCCEUYJlgIIYQQQhGGCRZCCCGEUIRhgoUQQgghFGGYYCGEEEIIRRgmWAghhBBCEYYJFkIIIYRQhGGChRBCCCEUYZhgIYQQQghFGCZYCCGEEEIRhgkWQgghhFCEYYKFEEIIIRRhmGAhhBBCCEUYJlgIIYQQQhGGCRZCCCGEUIRhgoUQQgghFGGYYHWq5cuXUxRF03RVVVX4+g0bNlAURVHU1q1b2/CwV1999Q033NCSLRcvXqxWqy90r9frfeutt/Lz8zUaTVpa2o033rh79+42xINQBHVQq4EINZzmW82cOXNSU1PbFh5CKK5hghUFcrl87dq14Ws++ugjpVIZrXhCPv300zNnzrzxxhu1tbVFRUUymaygoGDPnj3RjgshbDUIoTiDCVYUTJ48+aOPPgotejyer776asqUKVEMSTBz5sx//etfQ4YMUalUubm5K1as4DhuzZo10Y4LIWw1CKE4gwlWFNx5550lJSXHjx8XFtetWwcA539V/Pjjj6NHj1YqlVqtdvLkyaHtAeCzzz7r16+fQqHIz88vKSk5/ykOHz48ZcoUvV6vUChGjhy5ffv2NsRJ0zRFUaGTBO+9955wRYam6ZycnBkzZlRUVLThYRFqg/a3GrhYw+mIViOorq6+9tprVSpVTk7O0qVL2/CwCKG4gwlWFFxxxRX9+vUL/Rz/+OOPp0yZolKpwrf58ccfJ06cOGTIkLKysn379vl8vlGjRpWXlwPAtm3bbr/99ttuu62iouL999+fP3++w+EI3/fgwYMjRoxQqVT79u2rqqqaNGnShAkT9u3b1/IICSHl5eUPPPBAUlLSrFmzhJX3338/IYQQEggEvvvuu5qamhtuuIFl2XYdC4Rapp2tBi7WcDqo1QjrH3nkkaeffrqqquqZZ5556qmn3n333bYfCIRQvCCoE73++usAUFFRsWjRoh49ehBCTCaTWCz+5ptvvv76awDYsmWLsGV+fv7AgQNDO5rNZqVS+fDDDxNCxo4dO3LkyNBdJ0+epCjq+uuvD6255ppr+vXrFwwGQ2sKCgqmTJlCCFm0aJFKpWo+yFGjRgnvjdTU1B07dlxosxMnTgBAcXFxaw4AQq0WkVZDLtZwmmk1pAUN50Kt5sknnwSAr776KrTmvvvuS09PD38ihFCXhGewouPOO+8sKyvbtWvXp59+qtPpJk6cGH6v3+8vLi4OH99kNBpHjRq1detWQsju3buvvfba0F29evXq1atXaJFhmC1bttxwww1isTi0cuzYsUVFRS2MraioiGXZU6dOjR8//uqrr962bZuwPhAILFmypH///hqNhqKoPn36AEBZWVnrXz1CbdHmVgMAzTecjms1AEBRVPjzXn/99dXV1WfPnm3x60YIxSVMsKKjR48e+fn5H3300UcfffTHP/4x/GMdAOx2O8/zycnJ4StTUlIsFovVag0EAo3uCl+02WzBYPCll16iwixatMhqtbY8PJqme/bs+eGHH6akpDz77LPCyjlz5rzyyisvvfRSZWUlz/OVlZUAEAwGW/vaEWqbNrcaAGi+4XRcqwEArVYrlUobPWmjkhMIoa4HE6yomTZt2po1a3bt2jVt2rRGd+l0OpFIZDKZwlfW19cbDAaDwSCTyc6/K3Rbq9XSNP3cc881OlfJ83xrIxSLxX379g391F67du3DDz983XXXabVaiqJCXVsQ6jRtazUA0HzD6bhWAwAOh4NhmEZPmpGR0dpHRgjFF0ywombq1KkejycnJ2fkyJGN7pLL5cOGDfv2229Da6xW644dO6688kqKooYPH75x48bQXadOnTp9+nT4vldeeeW6des4jmtnhIFAoLS0tHfv3gBACPF6vTKZLHTvJ5980s7HR6i12tZqAKD5htNBrUZACAl/3u+++y4tLa1bt27tfCKEUIzDBCtqkpKSgsHgmTNnKIo6/96FCxcePnx49uzZdXV1ZWVld9xxh0QieeqppwBgwYIFO3bsePbZZ81m87Fjx5588snBgweH7/vKK6+cPHly2rRpx44d8/l8J06cWLZsmdDZtnn33nvvxx9/XF5e7vP5Dh06NHXqVJPJJFzsoChq0qRJ77zzzoEDB+x2+5tvvlldXR2hI4FQS7W51cDFGk5HtBpBcnLyqlWrduzY4XA43n///Q8//PDZZ59tdH0TIdT1YIIVoyZMmLBhw4bi4uJu3boNGjRIKpXu2LEjJycHAMaNG7d27dr//ve/mZmZM2bMePbZZ7Vabfi+eXl5QiHpsWPHGgyGm266qbKysiVfFfPmzdu+ffu4ceMMBsMtt9yi0WhKSkpCw6PefvvtESNGjBs3Ljc3t7S09LXXXuuA141Q2zXTauBiDaeDWg0AUBS1bNmyhQsXknVKoAAAIABJREFUpqWlLViwYMmSJQ888ECkXzpCKOZQhJBox4AQQggh1KXgGSyEEEIIoQjDBAshhBBCKMIwwUIIIYQQijBMsBBCCCGEIgwTLIQQQgihCMMECyGEEEIowjDBQgghhBCKMEywLqimpia+ZjL2+/11dXXRjqJ1zGaz1+uNdhSoAzEMU1tbG+0omkAIqaioiHYUTauurmZZNtpRIITaBRMshBBCCKEIwwQLIYQQQijCMMFCCCGEEIowTLAQAABcey307w//+U+040DokhDkf5sElgCwPM4Ji1BXI452ACg2nDwJZWVgMkU7DoS6viNmV7XLPybLIBfTBGB/rd0RYEdnGqR09H7xbt8O585BTg4UFEQtBoS6FjyDhRBCnYcQcAZYF8MWVlj9LLe/1n7O4QuwfDC6J7Fefx1mzIDXXotmDAh1LXgG6xLm9cKDDzbcrq8HAPj8czh5EgBgyBB49NGoBYZQ10VRMDxdV1xtr3b7N50xsTxRiOmCbKNKQkc7NIRQJGGCdQkLBGDVqt+tKSmBkhIAAKcTEyyEOoiIooal674vM/lYDgCGpeuill398gswDACAy9Xw99gxAACJBHr2jE5ICHUVmGBdwiQSmDSp4fa2beDzQa9ekJsLADB4cBTjQqhrIwAH6pw+lqMoIARKah1jsowycTQ6bFx9NZw9+9vi5s1w2WUAANnZcO5cFOJBqAvBBOsSplbDhg0Nt3NzoawM7r4bnn46qjEh1PUdrHOcdXgVYnp0luGIyVXt9m+vsIzJMsqjkmMhhDoGtmeEEOo8BIAjDf2uNFLxsHRduloetWg2bIADB+DAAZgwAQDgqqsaFjdujFpICHUVeAar63Ayv5u8zBEIamWSlu5sNILTCSpV5MNCCIWhAAan6gIsL5yvEvpjBXkii0qNhr59G25otQ1/8/KiEAZCXREmWF1ErTuwu9adIoUUAAA46/CW1Dr6JWr6GtUt2r+4uCOjQwj9hgIIvxoooigZTUUxHoRQR8AEq4tQSWmJCGoC5ECdQyeXlNQ6KIrSyvD/ixBqgQcfhEmToHv3aMeBUNeBX8BdhEYqzk/R7Kp1ldm9ACCiqOHpurQo9u1ACMWR8eOjHQFCXQ12cu86VBI6TdpwO1klxewKIYQQihZMsGLXoXqnycsItwmB/bUOmz/YzPZVHqbcT0QUJaNFte7AwXpnp4SJEEIIocYwwYpRJi9z2ubZWWWt9wYIgT01trMOb0md40LbV7r8h8weABierivINsrEol9sniNmVyeGjBBCCKEGmGDFqCSldFCKluPJrkpbUaWl0uVXSegRGfoLbW9USFQSuq+KSlPLNVJxQZZRLRUnK2WdGTNCCCGEBNjJPXb10CkJIQfrnSYvo5TQBdlGhfiCE5YpxPSYtASns+EUl0YqntAticKh3wghhFA04Bms2EUAQp2uAizv+n0d0fM1Sqcwu0IIIYSiBROsGEUA9tbYy50+lYTul6jhCNlVZQv1eUcINVLl8gd5PrRY4fRxPIliPAihSxwmWDHK4mMqnT6VhC7INvYzqvOSEzieHG1/p/Uvv4TLL4ehQyMRI0Kxosbt/7naVlRhDXI8ABy3uPfU2Itr7NGOCyF06YqhBKu8vHzevHkDBgxQqVQ9e/Z89NFHLRZLtIOKmkSFdHiGPtTvKlevGpqma6aTe0vZbHDkCBw5EoEQEYoZyUpZklJq8wcLK61Hza6jZpdYRPU24NyaCKGoiaEEa8aMGV999dXLL79cX1//ySef/PTTTyNHjvR6vdGOK2rS1fLwXu1ZCQppVKaDRSjm0SJqZKYhWSmz+4PHLW6xiBqZYTAqpBffs/OxLPj9EGyuph1CqAuIoS/sCRMm7NmzZ+LEiSqVaujQoW+//fbJkyc/++yzaMcVfwiAM/Bbj3ieEP+KlTBzJsycCR98AADAMA2LM2dCdXXUAkUocmiKSlQ2ZFRKCR27E3E++CAoFDBpUrTjQAh1rBj6DHr66afDF3v06AEA586di1I4cayk1lHh8o3MMCQppTwhP1fb037Y1u3jD3/bguPgw18Xn3oK0tM7Iowgx0t+PeVGAFieSEQ4shF1lJNWt3BlMEEqtvqDRZXW0VkGiSiGfkMihC4pMZRgNbJx40b4Nc1CraKVic86yM4qa366vszurXH79X368tdcI6IoqKqC0lIQiWDChIat1eqLPqDFx5ywuAen6uRiEQDUegIVTl9eslZKXzBhKjW5qt3+MVkGhZgmAPtq7I4AOybLgFc5UUeodvlLTS6xiBqVadDJJDurrCYvs7fGEYFui+HPUl29adMmrVZ73XXXyWRYxRch1ByKkFgcyWw2m/Py8miaPnHihEKhiEoMNTU1iYmJEokkKs/eBn6/3+FwpKSkEIBzDu/+2oaio2qpuCDLIBe6c33wAdx3HygU0JrObdvLLWYfI6JgdKaR4fniahshIKFFBVnGhKYuxBCAPdW2SpdfLRWPzjQcNbvKnT6lhC7IMiolv6uVajablUqlUqls+8tGsY1hGKvVmpqa2qHPwhMoqbN30yqFflccT/bV2vsaNU2+PwWEkMrKyqysrBY+xa5duyZOnOhyuQBg8ODB27dvV6la04n+j3+EY8cAAKqrwWYDlQq6dQMA0Ghg165G21ZXVycnJ4vFsfsDGCF0UbHYgIPB4NSpU00m06ZNm0LZ1fr169evXx++2eTJk0ePHt1xYdjtdpFIFEcJViAQcDqdYonkuIsVAajFlJslAAAcd7TW2k1JA4DM41EBEEJsVmvLH7mvEooDwPBQWGEBCggBCoDl+VqbnZU2fUYqVwYMQ9cH2E1l9TyAnKYGaGi/y+H//WY2my0QCPj9/iYfpCMYDIZOey7UaUQUDEnVhRZpETUsPZLnrgBg4cKFQnYFAPv37//444///Oc/t2L/06d/N3rX42lYTEiIYJAIodgRcwkWIeTuu+/esmXLmjVrrrzyytD6zMzM/Pz88C0zMzOl0g4cJSSRSKRSaRwlWDzPSyQSXiS2B5kARwBARlMBjrg5wgf43AQ5LaJolYoYDKBUtuTQ8YScdQe7qSVSihqdIi2q8zA8AQIUAEXBAL0iRdHc+2egVLKj3utjeQAYYFBopU3M8yMcYVltLeXxEJ2OpKW19dUj1LF8Pl8zixf35z83DCj59ls4cAC6dYNp0wAA5PJIRYgQiikxl2A98cQTn3zyyfLly++8887w9YMHDx48eHBnRqJSqdRqdRwlWGKxmOO4JF2Czs3WeQIAQFEiAA4AGJ5wUrlWLoF774V77wWAi3e8AjhU7zztDLg5yM/Q13oCzK91sQlAH4M6N1HTzL4EYH+tw8fyIoriCTnmYIT+WI028/v9SqVSOX06bN4M99zTMMgRodjzxBNPFBUVcRwHAJmZmXfccUfr9n/ooYYbdXVw4AD06AGLF0c6RoRQDImtHsdLlixZtmzZ4sWLH3744WjHEsdytEq1lKYA/CwnoUUDkhLUUrFacsGJos8X5Hk/y/cxqhNk4lpPYMs5y+6q3xXFPmlxW3zNzdtzoM5xzuFViOmruyVmJSjcDFtYYfWzXBtfErowrNDbOSZPnrxr1665c+e+8MILe/fuTUpKinZEzamoqJgzZ860adM+/PDD2Oxoi1CXF0NnsFasWDFv3rw5c+Y0qteAWitdLSt3StwMBwBBjncH2SuzE1teIYHlyY4Ka5AnY7IMY7OM2yusjkBDUUQKoODLNdIPVzp69Nyx9J0x2Ua9vOkzfCKKUojpgmyjSkL/IU0HAHZ/EADLNETejBkz6uvrly5dOnr06KNHj95zzz3ff/99SUkJDh2IuKFDhw5t/zRTl18OEyfCFVdEIqKmORyOkSNHVlZWAsDHH39cXl4+f/78jns6hFCTYugM1iuvvAIAL7/8MhXmr3/9a7TjijNC4atat18rk4zNNsrF9Bm7t9TkbPkj0CJKJRW7GHZ7hdUZYD3BhpqlFFBD03RGm1lz/Kjh3BlCAXvhyXTzkhPGd0tUSWgAoAD+kKobm20Uqjz8JiPDmJmpNBrhxx8BAFatAoUCFAoYNqz1r/vShRV648xjj8HGjbBkScc9w8aNG4XsSvDee+913HMhhC4khs5glZaWRjuEriDA8c5AUCuTCEWnxmQZtldY7P4gRwhNtegEkpAP8YRUufzbKiwAIKEphVjkDHDlLl8GAAWglNATuiUpm73sKAsreUVR0EQFLJ+PCgR+W+Q44DgAgE4cVNgFYIVe1EhKSkr4YnJycrQiQehSFkMJFooI4cIcTVFCQqORisdmGeViuvnsiiekuNqeoZFnJSgAgCOE4XiKAkIAgCrISlT9e1n93hKG45njpTIAKC9XzvpLw87vvQctS90aW77cbbFIpVLp66/DkSMwerTQAR+wkkI7YIVeNHbs2JkzZ65cuRIAdDrd0qVLox0RQpeiGC00Ggvit9BoG/Y1eZmiSgsADEvTp6hkhRVWm58BAJqiOEI0UvFV900VbdnS9M4sC3QretCHayg0etNNOIowImKhQm8jnVNotA2CHF9ZVd09O1NYFH5RxMjUOhEpNHr48OHq6urhw4frdLqLb40QijQ8gxVPeEJO2Ty99CoRRQEAy5Myu6eXQd3a00c8IaKwc048IUlK6dA03Z4ae3GNTS0RuxgWAJJVsvx0/d5ae7XLX97rsmyRSERRcPo0nD0LGg0MH96wf9tOX6FIa7JC79q1az/55JPwze64446rrrqq06JiGMZut9NtTcE7CEeg1MX5GCKtM0lFwBI45OREFFyuocWtfDvT586JqqqIRsMOGBCp8CwWC0VR7UywUlNTU1NTg8GgyWRqZzwxPmQSodiECVY8OWJ2nbJ6LF4mP0PPE9hZZTV7GY5AP2NLylo1cDNsUaV1YHJCuloOAPZAcFel7Q9pukyNguNhX63dxbBSkUirkIzI0NMUNTxdv7fGbnlucY5QKftvf4N//Qtyc2HzZpYnZXZvL1HDCEWG48udvp761swfIpg8GXr1gpEjW70j+tWFKvReccUVjU5l9e3bV92CCSgjhWGYYDDYmc/YEjwBhd/jZEmph1yRqCw1ez0cSVJINGplC7sqhkhWrRK//jo/ZEigsDBS4QlF+MRisTvIsQR0vxbpdTIcRYGmNSVXEELRgglWfLD5g3q5pLdBXecJ1HoCRRVWoMDsZfRySU9960bj2/xBX5ArrrYPT9dxPJTUOYI8b/ExermkzO4RtgnyfI5GIXzTCH3egWqixEJJnaPC6XMx7OBUbZDjCyusjkBQLKK6aVtZIADHirbbhSr09unTp0+fPtGKCgBomvb5fDFyvTJcfqZ8+5kaW5DbUeMiACkqWX6GvrXZFQCAWAwAIpEo9BpL6hwSkejyJA0A8ASKq20GhaS3oRUpplwuVygUhBLtqTZzPD8q02hQSOz+YLHJKaLg6pwkmTgmLmUihJqBCVYcOGp2nbC4B6dqvUFORouktMjsYwBARosUYlrcyl4jWQkKArCvxv5ztZ0nBAD6GNQ99apt5RZHICilRZcnaUpqHfvr7GIRla6RQ6NrgEol6HTCBGqXJWosPuacw8sT4gywjkAwWSUTusmjzoQVettAREGOhHfxYpYnIooalKJtXXbl9zcMemVZAACeB48HADwsX+XyCz26+hk1xdW2arffxbC5elWrz43Rolyd8ojZVVRpGZCUUGp2BTm+j1GN2RVCcQEbaszxs/xJqzu06GbYAMtTFLW/1nHW4TV5GYbjhbsCHG/xM60qjy6MaMhOUPQ1qvlfxzdY/cwZh9cZCAJAL4O6m1Y5NF1HCJxzNjXb2rPPgs0G27YBgEpCF2QZVRK6wukTsqtRi+fRkybBypVteeWoTbBCb9swHH8yQLM8kdIinpCdldYAy7di/2uuAbUa1Gp4/XUAgH37hEWVXjssQyelRScs7h/Omqrdfo1UPCbL0JZzYwB9jOoBSQksT0rqHEGO721Q9W92iiqEUOzABCvmFFfbSk2uQ/VOABBmmDnj8PY1qEUU5fv1CyD0Ua2RimXn15c6DyHkP//5z8xZD31YWHLo+Em7P/iL3Ru61+RlDtc7CUDfRE0fgwoAMjWK0VnG4ekXH3wkFlGhU2hysSi4vRA2bYJTpwAgyPNnwp4FdRCs0NsGQZ4UVdq8PKSoZBN7JKep5S6GLay0hH69tMehOufQdB0F4GZYANDIxPLzZuFsuWSVVCyiAEBEUUK/SYRQXMAEK+YMStHKxKLTNs++WkdhhdXHct20yj6J6iSlNLQNAaAAVFLa7GVOWj0XeiiOJ0GeAMDzzz//wAMPyNNyEpJT91mZ7eVm4RJGOLVU3Cesf3qSUiq62G9uhuOLKqyOQDBJKVWI6XKHz/9rChjk+aIKa0md46wDc6yOVVpaSs6zfPnyaMcV0yQiyqiQaGmSn6GXiKjh6fp0tTxBKm5UpqG2tvb222/Pysq68sor9+3b97uHeOEFWL8e1q+HKVMAAHr1gvXrD7+3ZudbK50BtrjaFqp/w7ejFI4jECyssLI8SVRIeUJ2VFpt/mCbHw0h1JmwD1bMSZCJC7KMhRXWcw4vAORolVekaktqHbWegFws8rMEoOEDWykWa2USR4Btsko7R8iuKhvD86MzDatXr1ZqErr3H2g3mXRJSSwBoYhorl51xu7lCZHSIjfD7qqyjchsRT/fA3VOeyCYIhePfPtVliflDq/MYgIA9+afLA5vOsun9+qVORt7BaFYNDA5oZJxCu92EQXD0nUU1fitf8cdd9hZqDebKysrJ02aVFZW5qGkySrZaZuH9BnYRxi9u2ULAAQ0CbIbb8zwMUWVVuBJkGtIqkQUVesOHDO7+rX+0p4wcITh+D5Gdf9EzXGL+6jZVVRhvaZHUkvOWyOEogsTrFhEUxT96ye9REQdqneedXgVYjpJKS13+gCAAlCIaZM3QAFFgKSpZdnndy0nQIDY/cGiSutlgwbfs2h6rwGDju3f88vRQ0PGXkUI5GiVQv/0PkZ1b72qsNJa7w3srrKNymxpIfWByQlSmhqglVHPPy8ByP11vXrPbvWe3QBAJlxDPYnXqlAson4/9/j552uDwaBHJH36nfePFO9a8uDdZrN518mzbnVihkZe5wkIXeN7GVQmL5ME4Gc5micGhXRomn53lVVIryQi0bB0XXGNvdYT6G1Ut6GTe/9EjY/lLkvUAEBfoxoAJCIKsyuE4gImWDHHE+S2V1h8QS5DIzd7mdM2T4ZGliAVD03TldQ5ZGKRhBK5g6yYooTsqpde1UR2BUCLqBEZhl1VVpOXueu5lwlQR/bs+uyt1555Z5WwQb0noJVJUtWyPgY1AIzJMuyotLZqDKD8bNmgdeuA46B//4ahhqdOQSAQMBgDxiQJTSl6dI/AEUEoGiQSCXicZ44fGTBi9Ny3PzxavNOtTqRFVA+dqqdetaPSetjkrHb7xUNH2SVyQ046RXgC9FmHN3RFMMjzpSbXyAx9gkzcTHZFyO8G6oZfUeyu+13Fk76tqXiHEIounCrngqI1Vc4PZ0xOhk1RykZmGVwBdlu5OciTAckJvfSqIE8CLCcX00WVFqsvCAA9dKpBKQnCjk1OleNn+e/L6jlCRBSV4Kx1apJ5oAYmJ7gY9ozdK0xcqPq1biGBJopdNefLL+GWW4TnBpksyBPmsv6qE8fK/zq79PG/+1muh045KEXbzAM0TJWjbGXdLBQ/YnaqHEJIZWVlVlZWM9vs27fv3j//5c6/L+yVNxgafrTok5UyALD4mMIKC08AeG7xvX8q2b0jOzv7P19u8Kj0AJCskll9DMsTAMhOUPwh7YLjRdwMu7PKNiglQXhYRyC4u8rWXcr1SE1qZyV3hFB04anmmCOcQzL7mHpPwB4IsjyhAIQcSCKi1FKxO8i6mYbSDPYAI3RjFzAEwks82PzBrefMHCG0iOIJCejThOyqp141KEWbo1X6Wc4e1me2nVPeHKhzcDwBgEyNoiDLIBfTZXYvdnJHceqbb765+uqrD5Xs37/tB2FNglRslDeMNfnF7mloeSK6+8DB+ROuk+sTj1XXAUBWgmJUpmF0pkEY/ZeR0NzQP4uPcTPsrkqb0N4LK6yeIGdnO/KFIYQ6Bf5Cijl9jGopLSqpc+yqshFCKAqGpelDw7Pt/mBhpTXj04/zXl7EJGg3fL9zZ6V11K8f5Sc8xOlw+Vl+YHKCzR/cWm4hhGhl4tFZxp+rbWYvkyATd9MqHYEgx5PBqdruWoVBIbX4GIlIlCBr8Zvh7Fk4dw4A4MiRhjWFhSCRDOR4QhEAEFGglooLsgynbJ4mL18iFOMYhpk5c6bdbr922j23P/J/Ab/PoFLa/EFhIMihOmel0w8AOt5X6+enPTGX5zmvy/X8rLtWffXtH9J0FIBBIR2daajzMmmq8xKsTz+Fb76Bbt1g4cIcrZIjcKDOsavKJhJRQY7vbVAbgq6GLd1uePFFAID77oNu3f6fvfOOj6pK//9zbpneZ9ILISEECKHXUAVEEFlR1EXxq2LBtiqLbV3XgrqrAu4uYvspoMiqqyKuKIhUIfTeQgklJJM6vc/cen5/3GQYQoBAghvd+37xx9w755x7cpg789znec7n+QUXQEZGprXIBlZ7RHTXKiMcozECQGeLTpJTlzjhDXOCmE5h0utVA1jVCneUrQ8zGXoVAOSp0ZEIOukNc6JYH2YxxiqKGJFtowg0JMOyucpNEwQniiV2j4hxvzRTuk7lirJbqzwkQtd2TFK0MHn2o4/gb38758y11wKAEgB+/hkGDZKKh+gUVO+LxgdlZFpClBfUCTpSTQ6vEj6fz+12Dxo74b6/vMYysdcfuvuLRR/Va2yOCLOzxueOsjSBOBEHSc3hrT/2G3MdgYjFc16dFPQPeO0F6NABnn8eACxqhUWtaGb0PXtgyRLo1QteeQUAck2aKC8cd4cEAWfq1d2T9DU1CQbWa68BAIweLRtYMjK/LuQQYbvj2WefnTbj6ahSJwg8AJz0hOvDTPzdvqnG/mmm1EaH1pBMy8B0c0ajBaYhYXi2VUURFf5ojBc6GDVjOyZTBMIAe+v9PoYXRIwxdLZoeRFvr/auOePcWuXhRZxv0dIkURU8K90uYlwdjF327CkKlEog5WK0Mm2DL8atPeM66m4IfNeHmdWnnae8F9R+ayuSk5NHjBixb9P6vRvXvf7Q3SgS6FrQeVimJVmj7GrVDcuyjMi2FZpVvCD0GzNOFPGnc17VseG/jBkDH30Ey5Zd1rX8DBeX5K0NxRwR5uLtZWRkfhXIHqz2xeHDhzcfOjZj7jsYi2/98aHc/M63Pv7s9mrv4ExzskYJf/878dNPWQBgtwMABIPU+HEZUs/Fi8FkgobN5w3JVO4oSyIkVR60B6IIwBllf65wx5NCggwPABl6VWeL7qAjcNIbrjHEsvTqFK1yR42vNhTrk2pspnLzc881lGf+8Ue47z4AgDNnQKEAALDZrur6yPwPggCOuoKCiJM0iu3VXuGX2pezbNmyl1566edPPxjYvdurr75KEAQBMDTrrIjJ52vXWHsOIglC4Lgzx48O6NZZp7voLr9vv224cyXNUqcT3n47ygtVvgh7x7R8q05HU1JuQGHIhbZvB5IEn6+h75Yt4PcDAHTvDnl5F7iAjIxMO0LeRXhBfuFdhDzP//3vf1+yZIk+I+exN+fNe/oPO9b8qNFoSmucBx2BwRkWq5r23vF/1n9/1nz/U6di6el1Ht/RKBHlhTSdqi4UwwB6BWVS0fZAVEOTOUbNEVdD6AGh+H89QoD7pJrMKnqj3cUJGAFK0iocYUavoIZlWVUXqSx77i7CK/ir5V2Ev3lav4vQG+M22z2cKErquFkGtZTkBACCiO3BaDPPAC2gJbsIL0JNKLbN7sIYNn2/bIrRRNdUlW3fNBVhWL0asrMhXhdy+vSzfUaPhvXrmx2ttNZbmGoCgNO+yP56f88ta/Luu7v5C8+ZA089dWVzlpGR+SWRPVjthRdffPH1118HADh8+OjuHT63EwDuvffeHKMmTatSUkS5L+It6hvl+BStkiorQ9u3izTtuflWm5TkodcDwMkoRHmho0nTO8XojLCb7e4gywdZXkOTw7KsWpr0RNm6MAMAknWVrFE6IgwG2Fvny7dopV1RGHCjdWW5mHUlI/OLYFbRXW26g44AxqChSXsgSiDUJ9UoinhrtccZYXkRd0qo8tQsEU7Q0Gcj13vq/Om6hkcCDLCvzt/BqLY2my/VHDFe3FXjQ4j44MWn1i794v6U1B71df3jb1dWwoMPNrx+4IGzIldmMyQlAQCEwxCJAEWB2QwAGENhcoPYSq5JY1XTwo4WTkRGRqb9IhtY7YWvv/46/trndvbt2/fuu+9+5JFHAEBJEQDQ0aQJ3H/fTu8UFUV0+mJx5+3bRY3W+Pm/gGj8+o7FOmtQiNZKus82jSJZo6qPxADApKQ1NHnEFawLM5IbAAAoAnqnGqsC0VJXEACkmoZamgxzAgAkaRQ0QfAijoSCzz777Pr16zMzM1977bXBgwefnfTgwbB8OQDAL64WJvO/gyPClDqDAEARKMIJNElU+CMY4ygvOCOsSUlfcqeqO8puqfLkmrTdk/QAcNIbrvBHKv2RTkqUCbC71mcPRD0xdnROUguVSlQU0TfNhDHOff6ZD6wGvGIFr9GQBIlYBlgWCAK0jQZfooTo0qW1oViEE/LeeAVmz4bu3Ss3bgWAJvM3KumaUaP4QICiKKivh44dAQB++gmGDQOQ7zUZmV8NsoHVXsjNzT158mT8cO7cuSNHjowf8iLmRLFnsgFjfNoXiXACAFAEOmtdAQCAkoBsmx4AJKdUfSSmoUkEUBOKrTvjDDC8FBlUUUSMF3kRNlW4RubYIrxQ4Y9IVleYE3Q0xYniaV+kLsyoKXLhS08v+eRjACgrKxs3blxZWdlZLdPUVJg48SrMVDGzAAAgAElEQVQvjMz/NM4Iu63KK2DcK8VoVtGb7W5OEGkSSTWjjEp6aJblkrtfSYQIhCSJuCSNQjLXMMBJhvTZ3c4Iq6XJ4gzLZenAZepVAJDVq9cHH3xw9uzTT8PcudCrFzSpDA0AAJyId9f5OUG0xDgzACuIe2p9BEJSrfRzmhIEqNVAUaBq3EGsUIBaVjyRkfk1IQeA2gvz58/v1KkTABAE8dRTTzWxrrZUeTZVNsT7AIA1WQL5XfguXQEgygv76v3xzN8gy++v9x91Biv8UQ1NDs+ySlrtAYYnEMIYJ2kU13VM7p1qpAgUFcRDjoA9EBUx6BQNX/GciPulmQmEIpwQ4YSff/45PpNAILBly5Y2/KtNo0aps7Nh/vw2HFPmt4SKIhQk0TPZkGvSmFX0sCyrkiTUjdtUjUqKboG2iElFD8m00CRR5gltrfKIGPdJNfZJMWAAZ4RVU+SwLGtiAPEqQRNocLqZIpAzwgBAlBcQgv7ppl9AdUJGRuaXR/ZgtT0Y48OHDwNA9+7dUYvLu3bu3PnYsWPHjh1LSkpKTk5OfIskkJYm3VF23RmXiDGBoOr631Vd/zsVRQxn+VJXsDoYC3NCb6s6KsIeuyfGCz2TDWk6Vc9kg/SzMTzbur8+0CPZUB9mOhjUJIE6GjWZevUZf/iEJ8yLuHuSPkOv3l7jDcR4RhC2VnswxhSJhmdbc7Iy7WfKpZkQBFFYWNiGa0V4vcjthois9v4/gcfjeeONN0pLS/v06fPss89eYs8dAADoFdSYjkl0o6dWr6T0KtoVZoxKmhXEykCUQKh3qvGSt5lZRXex6g45AhggTafqYNTsqW3YoMcIoifGnvbxXW0N9ZhZQTzhDXez6i+zOvOlsWkUxZmWnQ89cXrqNKygB6Sf1RBuHqsVNm0CAOjRo42nIiMjc5WRPVhtgyiKTqeT5/lAIDB06NAePXr06NFj6NChgUDg/MbBYPDee+8dPn5i165dv/rqKwDAAIu//nb06NEzZ87ctWtXk/YIoG+ayaKmRYwBQMRwZOdWwu+I8WKZJ9w7xWhS0o4ws8sROhzCMV7oaNLkmbWDM8zxh3I1RQ7OMGtpMtekIRt/q2gC5Zt1A9LNPZINnS06LU2O7mDrn24kAGGMsSju/GJh1O9duHBhz549AcBgMLz//vsFBQVXbxnbhvXrYckS2Lr1vz0PmXPAGI8fP37OnDkrV6587bXXbrvtthZ2pBPi4AcdAVeYMSnpYVmW4dlWNUWe8UdaIotVH2ZKnUEEQBGoNhRbf8ZVGYgqES5M0mOMd9X6yjwhSQOCFcQSu+e4O3TCG7rksE159lk4fhy+/fYiTWK8yOj1kbSMqDU5xF6qJg5Nw7BhMGwYGI2wYgUMHw7XXSeK4pw5c3r16jVkyJClS5de9iRlZGR+EWSZhgvScpmG7du333HHHeXl5UlJSaNGjfryyy/jb7300ksvv/xyk/YPPPAAY0qddP8ji998ZcWnH+3bv/+In6UsKf946pEtK5eTJLlp06bi4uImvSIs9/3xalqhFAR+xoSRQbdz+/HyLqkWAiHpJ8HPcACQY9T0Sb1C/XQB423VXkeYYWMxhUpVUXb069kvl/y8gSRJl8tlMBgUivO2WYkiYAwEAS1/2Pf7YcoU6SXesAExDBQUNGTyDhsGf/7zlU3+LOPHw6pVcNddsHhxa4eSaTVxmYajR49269Yt8a2ampq0tLTLGi3GiwcdgV4pBinvKswJR13B3ilGkrjYx68+zEiFp3qnGo1KuqTSw2NRQaCuSj4vO/OMP7Kvzk8gJGCcrFEyguhnuBStclCGmWyFC4sXcYU/kqxV6hWUdHjKG6ZIdLA+gBAUWHRl3rAg4p7JhrzztkDW1NQkJyc3Lfa8cCHcfz9oNPP+9rcZM2ZI5wiC2Lhx49ChQ694njIyMlcJ2YPVWkRRnDJlSnl5OQA4nc5vvvkm8V27pCt4Lps2bdqzcW0sEr772RcnTnvwkCdK29KctdVl+/cCgCAI3573BMwKYkmlm1YoAx43SVLPf/gvhVbHO6uJxogGK4hSywgvXFKJkRdxVYJKOyuItaFY3Lo6feTQI9cO3rtpfYfOXW956qV9Bw8BgM1ma8a6AoDcXKCohnJpLYRlYdUq6R9iGACA48cbzuzbdxnjyPyqSE5ObvIRuv322zmOu1D7ZlFRxIB0UzyrXUuT/dJMF7euAEBDk0qS6JlizDFqzCp6eLaFJlG3JL0CYQDIMWoGpJvHdLSZVbQjwrSJdRXlhdXlzgOOQIndHWT5EMuvPFVf6goedAQRgkKbwaJWFGdYSAJJIX5P9DLWYcWKFfHXoiiuXLnyiucpIyNz9ZANrNYSDAYrpMrHAADA83w874ogiNtvv/38LgMGDDiya/ur998RDYf+76m/qFIymVDgpbtvddZUSQ3y8/MT2/MiLrF7wgI+c3j/Q6MHbl7xXWp2zt8+/y63cwEAhFi+xO6J8kKmTqklkSPMXFLtemeNd2eN96Q3DJLpZvdsr/Y6I6yKJAwK6t2nH/U66998dNq+kg2xcCilMSGME9vI2alSwV13Sf+wJDHap0/DmYTU/svmxAk4ehSOHoVwGADA7284PH26TWYt00qsVuvbb7+dmJW4cePGVatW/QKX1iuoazsm5Zoa9EhNKnp8bkrHBHnSDL0qcSti6z/oUU7kRREAYry4qdK9rsLJi5hEaECaqV+q6bgntL3aK2JcnGHpZjPsrvNtqXLHeLH5sb75BmbOhJkzQXKNs+yfXa6/A/wdQLoz2zYtUkZGpq2Qk9xbi9FoLC4u3tqY8ZOdnT1//vxPPvkEAB544IExY8ac32XevHmBQGDVTz+dPLC3qHg4AHROsXTOyXZUVQLAlClTpk2bltieIpBNo1BRxKAhvffefNOqhfOzsrJ69uyhVSoB4IgrJImLdjUq3T7uWIyqDzPVwdhFxIHyLTpXlD3oCPAirg7G/AyXrFEmaRQpWiUv4n/MnTN16tSiISMKevaFkJ/VGAHgjD9S6graVIo0veqSskOXQK+PB+/w+vUoEoHbboNnn23VmABQXAwu19nD776D774DAOjWDUpLWzu4TFvw4IMPzpkz59SpU/EzDPML1d2jzvVyUcQ52RHxIHuyVsnwoiPM7N1/vC/FERrNldWlsajpIZmWLVUeXsSMIAIAidA1HWwGJQUAnIj31fu3V3sLrLrjnpAg4m7JhguK+q5dC4lKEDw/ct++kQAAsIgkb3nwwWaf4mRkZP7ryDlYF6TlOVh1dXVPPvnkzp07u3btOmfOnJakgUv1ASsDUQ1NsoLIi7hHsoHw1qtUqvT09Ga7iBhLAUFGEEvsboOC7p9mQggCDL/J7s42qDsbFH6/32xLqgrG4g/rF0KSXuRFDAA2taI405L4C+T1eneergrrbIAAMGTpVVXBGCCQpLSvNdPkqFENTY8cAZaF9HSQHF1jxsCcOZf888/+UVlZRFUVvPFGGxhYSUnnGFhxZAPrv0qTUjkLFy584IEHpK+d7t2779y5U31F8k4xXlSQiGj0h0V5QUWRlxXSSyyVs6/eX+6LSJFBQcQldk/2Ky/kf/wB9O0Lu3dfwfQk7IHorsa9ikqSGJZtNSganmmlxC/py7cwSV9gOWdD5Tk5WG++CVLigcsF5eVAENC3LwDwPB/79FNd9+5XPD0ZGZmriuzBagNSU1M/+6z5EoHRaHTFihWhUGjs2LGJltNBR0CyroZnWSO8sLXKc9AR6JeWln5h51D8t0QQMS/iqmCUQNDVpt9a7ZFysKQvawVJXNK6AgCDgtLQZIDhASBVp2zyfG82mwd319uDkePuMCCwB2MIAGOQFINIvxf27z9nuJoaqKkBgMt93Bfy8kCnI9qkRPTPPwPPAwA8+ihs2QITJ8KrrwLAWalGmXbAfffdl5+fv3r16uTk5HvuuefKrKsoL2yqdOsUlJQpFWT5ErvHplZIjxxXQFGSQUEQXWy6M6dPL168WETEg/jyksPOJ8Ty++r98UNGEDfbPcOzLDoFBQA6BUUQSBAxgcCkvOhT3LPPNjx+SEnuKhXs3AkAFMClVS5kZGT+e8gG1lUkEAgMGTJE0sTS6/UzZ87897//7ff7J0+e/OLf3nCr6IHpZg1NamhySKbloCOQrGlRvWTJLNtkd1cGolXBmIhxtkHdN83ExGIX6VVWVrZ///4uXbr06NEjxgtbq70BhreoaD/LH3YGSYQStzKFOaHE7qZJIlWnrAsxACACEICGZFm0NAkaDTzzTEPTDz6AQABGjICBAwEAiooua4n8S5e2WbHneCaKXg8AYDZDz55tMKxMWzN8+PDhw4e3ZgQCIZJA9WFme7W3e5J+S5U3xgsUgeBKs9IpAhUm6cvKyvr37y9Jq9hIckYrZhhi+Z8r3VLeVe9U4/56Py/iGC/sqfOPyLa6ouzWKo8gYun+2lbtHZxhTtEqgyy/s8bXP83UiitfkPowo6FJfaMLrSYUkypoXY1rycjIgGxgXVWWLFkiWVcAEAwGZ82aJb1+9913NRrN7Nmz4y2tasU1Hc7x4rCCKGAcl3iO8gKBkLIxD1dDk/3STJsq3SLGGprsk2q6+C/LO++888QTT4iiCADvLf0+o0c/XsTJGuXgTPPOGl9dKHbAEdApqBRtg4WnocgkjbIqGA2z/OCH71Y56k//370Vk25zhhmDggKN5uy2wS+/hEAAxo2DP/2pNWslI9NylCRRpCdXVVcDpNSHYoBQrknTK+UK1UniLFiwYGkgIJl+lCAAAOzde9YDetEHmCZgAAKBWUX3STUZlZSWJrdUeYxKun+aiRPEbVVeXsRFyQYCoC7EiBhvr/EOybLsqPYyvFgVjDZvYSEECAFxJTuTAiy/vdpLk2holtWgoOyB6O5an1ZBjcmxEW2upiojIwMA8i7Cq0rswt/IifVnEnf8YQARY04US+yeTZVuqeagFBApsbvPajFwgqRDLRW02Vfvu0gmXTQanfv2O5J1RdG0K8LwIlaQRN8040FHoDYUo0mUoVcl+s8QgmStAgFgAOPxo+bSgyqHAxAccAQOOYPxaTSZfHvhj3+EBQvgvvv+2/OQuYo8eN+9bzw+HQAAIU99XZbiAlvwLgedTqcAUAIoARqebDAGhpH++WLc7lqf2LIPvF5BXdPBdk0Hm1FJAYBVrRiTkzQi26qhSZokBkTdo2e/mP/803kE39WmBwDAsKPKy/BinrmhWHsz3HsviCIEg1fwpxkUVI5JE+PFErv7uDu0u9YHCLrZ9LJ1JSNz9ZANrKvInXfeGc+7IslzXPH9+/eXXpS6gpvtDcnmALC/3l9i99SGGKOSCnPCJru7MhDZUOEKc4JBQdEEAQARTiixu8OckKlXX9sxSUuTFf7o3roL2lj1oejrX62YOvM5AOA5btXnnwAAK4jrzrjKfREVRYzItg1MN5//TZt4JkmjIBFCBDrpDZXYPZKNVRuK1fbs5ysehnNyWrFObc3YsXDffdC6IJRMO+fA0eNP/vNDAIgEA5aU1M1nHC00fS7Cww8/vCQjYxrANIANSiUARDKyjs+dLy5aFPlwweYqjxSUT+zS5KKJh00qDGrWroY334SvvgKAFL/buGgBvP8+hEJdrbo8s1bAmBHETL26Z7KhlX/FheiZbOhk1jK8WOoKAkC/VJNUr1pGRuYqIYcIryIpKSl79uz59NNPQ6HQpEmTNmzY8NprrwWDwZtuuun1118HAAFjZ4T1RNktVZ7iTMtBR6DCH0EAnijbJ82EEFT4o7tr/QCQolX2TzN7Gc6iokkCkQTK1KvzzFotTQ7LspbY3TRxQbFFrVYL2HPz9MdIijq8Y+uMue8BAE0QkpE0KN2iU1DuKGtVN4hAihj73b60556FUAxjUPq8AKD8Yfm4umoR46r8bocmTSmxezqZtfvq/XjOO31SjSZjWyRRyci0jBDLv7x4qd5s+emLxV+9+49XFn+dkZe/s8Y3MMPcGodMUlLS22Vlq1atYlm2eMsWeOcdwWIpvWFyjYoOcwIriAUWXaJGiTvK7q71DcowG5U0ANSFmYOOwJBMi7bZxKavv4aPP4Zrr4VzawQFWb660WirD8e8Mc6suvTO5SvDolaANwwANEmYrtpVZGRkJGSZhgvScpmGy4Ln+cQKGLyIt1R53FFWQRKsIKopssCmO1gfwBjnmbWSFigCNDY3yRlh9tb5u1h13Wx6ThBP+CLHXMFeKcZck4YVRAVJxGIxv9+fkpKSeLkgy5fY3UlqusITJGgFYAwIaWkyzAlSA7OKNiipSn+0d6opx6gWMd5R4/NW1V4/qHn1Qjx58vZ5H9WGGn4SeqcYO7Zg0+KFcLlcbZbkLtMuaSLT0CZwIl57orpkzU9vPDHdaDDMf/+D1EGjMvTqrtaL7qsTRVi3DgCgd2+w2RJlGprhqafgrbdw374/f7PKG+MAIN+sLTrXvXTEFTzmDilIYliWJcIJO2p8GOP+6aZMvbpxnuK2Km8Xmy5Zo4R774WPPw53LoCZT2rPnIY33gAAZvac4wLFi1g9/jrIyjrqDtEkMSzTEvE4mymV0zqqgrFdtV4ASNYo68OMiiKGZVnjOe8yMjJtjnx3/dI0+dKkCFScaVl92iGpERZnmo1KWkOR26sblNYpAvEi3mx390w2KkjimDuEAQiEjrmCNIGkx9BEEWpOxJ4oG09Xj/ECL4I9yKQYdc4IK4X9wpyAAMwqhZ9hvTEuxgsYYF+dD2NcF2ZqQzGzWhkaPIRASEOTsH07MAyT3QFnZasoAhUWZhvUkoGlpckMOcog84tDE2hsfsb4gvuevH2SwWCgaZoXMXWpgjkQjcLYsQAAy5fDxImXaDx0KMRiTEZm/FGkKhjtatNLVznlDQsYd7PpRQxlntAmu0cQMca4T+pZ6woA6kKMsGPHyUhYYdPra2pIAG3ZcXhoeryB8pmne0ivCv8DvbsBwFF3yBlhm9YmbDUBhpesq/5p5ky96oAjcMob3lrlubZj8iWXTUZG5sqQDazLo7q6etmyZQRB3HLLLU18RVcGBjjkCDCNaeP76wNDMi0UcbZYR89kgyPC2gPRfXXeXd/8q/OY3x0HAACaQEMyrZbz/PzbqjzuGNs/zZSpVzO8eKA+gDGmSMIZYaXLIQCEQE2RnhgLAAqSiPIiACCEJNkenYIanJej2rq5YcScHKioUD78kLRPsC7M7Kr2IgC9kgow/Ga7Z2iWJdHCk5H5BZDqD1qtVunw0tbV5TJpkn/8BCndMM+kLfeHo7y4ptx5bcekCn/kgCNAIpSpV3dP0oe5hhhfV6uug/EcWa8sgzrl5WcURy6sc6tSYal4EEUBQFebPkmrtKkVNTX+C3a5IgxKqqtVr1NQUt5Vz2QDgSBFq5StKxmZq4dsYLUIURSXLFny448/fvfdd9LewFmzZu3YsaNjx46tHPlAvf+MP6KmyOJM84H6gCvKbqx0RXkRY6ymSEYQ9tb5+6aZEMCKH3549bln7guz46dOA4B0vcqsPse6kmyyPLPWXcvuqvWxAj7tDQdY3qyifQwvtTEoqRArYIyjXINJJ2VidbXqnRHGFWUBoKNRo6KaV8fxxrjt1V6McZ9UY6ZBLRWH3lrtHZltbeU6yMhcLXi+YeddJNJwJhzGHq8zwqDGQj0Cxu4o20SIrswTlvKuCpP0WQb1xkpXlBd+rnQFGJ5EaHCmWUOTdSGmNtQwyClfJEOvlorhSOyv93cWcXM10hs5eRJlZCSesKkv1rw1dDk3flqUdLWy6WVkZCT+JxwPdXV1X3zxxbp16wRBuLIRHnvssXvuuefLL7+MKy84nc733nuvlRMTMQ5zgpoih2dbjUp6cKbFolYEWYEVxM4W3bi85AHpZoRgT53PLERmPXjXzQ8+Pn7qtEgwEAr4K/zRNeXO4+6QNBQv4p31QQcLGXrVgDQzAthf7w+wPO9zmYUoYAwAJIGCDJ+qVQIABhx3O+kVlI9lXVFWQ5MEQoedgcpA9OwstVrQakGhAACTks7Qq/qkGjsYNSRCgzPMqTplgaXNAxoyMm3Hli1gsYDFApmZDWduvx1ZLclZaeT89wBAwHhblXdLlacufE5hxD6pxr6ppsIkPQBY1PSIbBuJUIDhEcDgTHOyRnnEFdxR48UY9001FVh1rCCW2N2BxocZVhAdEXbdp9/8tP3wiq2H7BMmAYB7YDFbWwfLl8evggH21PlqGpMaBYy3VXsDguxZkpH51fPb92Bt2rRpwoQJoVAIAMaMGbNy5crLzVtnGObDDz9seXsMwPBC3AkkYsyJWNlcEI1AaHCGOcaLkp4yTaAhmRZnmPExnKSFk65TDUg3RzghQ01Onv6H22c8GwkGXrn/DpokX//38hDLl7qCBEIdTZotVR53jOco1B3AplGoSCLCiwCw7F+fTrjrfq3B0NWmT9cpS+weKX2KQEjyXSlJIsjyQZbXKajhWVZvjN1R49tT61MQRKpO6Z//rvH3v4ehQ2HUKAxQH2YSZaZJhIozLJe1mDIyV4MYL5AEQTdGvCKcoKQI8lIiTz4BhbzhumDMGWWNSrpJwJ1EKDHkF2A5SfUNA9SFmADDH3OHEEDvVFMHo3p/vR8AWBFHeEFyYilIYmCacQMnhDEmEYhKJQCIFO03mJNMZ28id4St9EftgeiAdHOKVil5hUMkdEqcCsbS9hSQZatkZH49/PYNrD//+c+SdQUAa9euXb58+eTJky9rBIqiNBqNVEAjTnJy8iOPPNJs+/31/tpQTNqhI2K8s8YXYPlhWRZ1c3G3hkRygGg0+p///MfhcIwbN65bQrnodF1DFvkdE8dXnz45/7k/Vp84tmDBghHZ1m3VXkYQDzkDp33hMCeYlFSBSpQeoyO8ePLQ/o7duk9++PGw379x6Wc3P/9UlBekcrh6JRXlGooXWtVKH8NGOEGnIFUUkaZTDc4wH3WHLGr6mDuU9u67cPwo/OlPeNSovXW+Cn+0Z7IhsaiOjMx/nRgvbrJ7SCwe/OGryvLT106YCB27G5RUcYaZQAh69IBVqwAAGAZuvBEAap5+rrygyKCk6lMzA46G+zpLr7pIKuFpX0QyobrZ9Ke84ZPesEVNEwiJGLsijC/GnvZFVBTRO9WUqj0bZwzzIgaMEAgYpFQrjHF9KJbUt29DAfKUFBtF9Usz7a717azxGpS0L8YZlXQ+zZ5z+VtugWXL4JZb4Ouv23jtZGRkrhq/fQMreK7wcRM7qSWQJPnWW2899NBDUoTxmmuumTx58kWS3BGApJg8JNNy1BWqCcV0CuriD57BYHDo0KEHDx4EAIVCsWzZsgkTJjRpc9OE8cFQaOgnC/Ly8vR6PQBcl5vsCDNbqjxhTtDQZP9kXSQY2FrlCTA8Dnj+MnVS/9HjZsx9R2swlO7YAvAUhZCKIs0quk+qeWu126pW1IYYDU30TrWV2N2p2gZLLkWrlDYhputU0hOzI8LY63wV/qiaIlN18rZBmfYFTSAlAe6YKGQVrFj0SeFNd1l4YcuqH744uveN11/Xm81w3XUAAOGw1N5WPLC0Z3E9y5OoIXVRr6DyGiPdEU4Ic0KSpiEXKsDyLC+WeUJS2zP+SL80844ajyfKUQRSkoQUT1eSxKB0S6KFxvCiTkFJmggAUHb3A5FJN6dkZzRoPXTrFm+ZZVCLGO+p8/tinE5BDcuyuOrrrvKaycjIXHV++wbWE088cV9j1ZQOHTpMmjTpCga5//77hw8fvnv37vz8/LgI+4XolWLEAOW+yPoKF8agV1DDsqwq6oIPxytXrpw6darP55MOWZZ99dVXJ0yYUF5e/sgjj+zcubNz587z58/v16+fXqfr1atXvKMg4jJPg3MuwgmVIcYGUGjTl/sj3TI65OXmbv1xOQD0HjrytonXAwBNEkOzrBSBEMDwLCuBUIFFVJAEBhiRZaUbfxxEEaMfVyKEDABiLAIA7IlTzPIfskmisGuuOi/5ChZQRubqQRLIe2B7qZ8tHFD8+r+/Rwit/frzD156BotiwO9fsmRJk/YYQEWRUV6QKiggBEGWL/dFOpm1gohL7O4YL0pZVgGWL7G7RREGZ5pFDPZApMIf3VXr5UVMIVScaan0R8/4IwBgVdMHnf4gww/JtFrUNMOLJXZ3iBdEEWtockCaea+i6AjL+/WqAbhpoE/EuLoxByvC8dJeExkZmV87/xNCo2vWrPn++++TkpKmT5/ecm2F1giNihivLndKlQSHZ1svsjMoGAxmZGQ0cbP169dv165dffr02bdvn3QmKSmpvLxcqz0bm4twwp46nzPCWtSKrlbdjhovL+IcNeqT3aDo6PV6Fy5cWFtbe/31148ePTrei0Aobu0FWV5FEYedwQDDD8m0UAQSMeyqdAzMucAqjRkDa9ZcwYJcCFlo9DfP1RAalaiurn733XcdDsfEiRO1Wu39f3j8rW/XIIII+X2PXjs4FPADgEqlWrt27ZAhQwAAOA5mzwaAo6OvP2o+Ox8sioggAGBwhjlNp5KigSRCRcmGo+4gw4vxStIYYGOFyxPjAKBfmskb4055wyqKkEqCmlS0L8ZRBOqfZip1BgMsTxNIQRLDs61qiozxwia7h0RoeLaFRCjCCToFBY0p9o4IY1TSOUb1QUcAIZSvEgvSk6kbb4TaWgCA8nLw+cBkAmnbckYGfP99m6+njIxM29KODCyHw/Hkk0+uWLGCZdkhQ4b8/e9/LyxsXkz8l+F8Awtj/O233x48eLCoqOjmm29GF0g4FTHsrPXWBGOSRqi0SbD56hkApaWl3bt3TzxDkuSXX345duxYg+GcfdQ7d+6MO89qgrFddT6EkEFBDcm0VAejBx1BDNhA4JF5aRcKR8Z4cWOli0BI8qhJT+dqikQA3hhnVSsGZ5j31vnrfMFJ3Ts0O0Jo+Ejdxg0XGP5KkA2s3zytNLAcDofBYFCpmgamPR5PUVFRTeQ54QMAACAASURBVE2NdPj+wo+1hQO0JrPf7TJabScO7J113+3RUBAASJJcs2bNNddcE+9bUV2zxRFVaTRsNEYqaJIkKSxYtZqBGWZJT+uMP7Kvzi99M3Y0aXqnGKWOJ73hg405WzRJcIIo6aGTCG2yuyOckKRROiMNWxFTtMq+qUacUJQwxgsEQjRBbKvxeKPcsCyrQUk5wszmKg+B0DXZVqOKrgxE99T6DBSM6JBE5eVBZWUzi5KTA+XlV7aeMjIyvxjtRaaB47ixY8eePHly7969FRUVVqt15MiRtdLTW7vh4Ycfnjx58qxZs2655Zbp06c32wZDg3WlV1BjOyblGDVRXthU6Y5wZxUiBIx31fr8DAcAWq1WrdM//fZHqdk50rvffPPN5MmTdTpdly5d4l2MRmNBQuY7I4iCiLGI8y3aqkB0b51fxLiLWd1VdzHhQAWJjEpaKp7jiDAldjfDi1a1YmiW1aJWuKPsqtOOmlBMq1HH3B7w+4+erg7kdwEAbsYfgw73T/tOrHt74QlPuBVLKCPTUioqKvr375+SkmK1WufMmdPk3e+++y5uXdnS0lVd+mpNZjIWJiuORJx1+T37vLDg82tuuk2l0QqCsGDBgsS+P+0/qtJogl6PQq0iSTISCvKITNOr4mqlVrWCIgkAQAAZjUmHknVFEmholqWDUc0JIkmgfmlmvYLS0OTwLGuSRtEjWR93DxdYdCqKjFtXUlxSQRIIgVFJM4JYYnd7ouxxTwgA9ApS2uySbVAXZ1o6qzAAwEMPwdNPw9NPQ+fOAAAFBQ2HDz7Y1ostIyPT9rQXA+vzzz8/cODAokWLcnJyrFbrhx9+yLLs+d+q/0WcTmeiWMPChQvr6ppJREUABgUlZaqqKLJPqjHXpFFSRKLSdFUgZg9ES+weH8NZkpJnffLloLETps58DgC0Wq0UzkMILV26tF+/fgDQqVOnb7/91mAwiKK4bt26xYsXE35n7xSjgPGuGt++ej+BUMXm1UO7duqUl/fII4+wbPM5HARCA9LN6XpVkOU32z0ML3Y0aXomG2gCFWeYaQLxIkYIBmeYVRYzGAwdMlNIigQAWqXUW03FaaYkCrK1V0sIUUYmkUcffXT37t0AEIlEnnnmmZ07dya+a7Gc1QfxuZyOitMpWqWo0mYMHj15YI9QfbXAsX94/Z9P/vP/AUBSUlJi353ffbXh2684vkGwKuhxm0Smo0kDL7wACIndupXY3Zwg2tQKDLCtxitVQYhwAkWgIZmWZI2yT6op26AmEVKSDfe1hiYHpJt31fpivCgV+NtW7ZGKGErsr/dLlRIAoKtVp1OQjCD+XOl2RliLih6ebYsnQaZoG0d97jmYPRtmzwbJyV1U1HD4pz+10RrLyMhcRdqLgbV8+fK8vLyuXbtKhzqdbtSoUd99991/d1aJEARBEGeXCyFEks1H/brZ9KM62OI6WD1TjMOzrIk7wDsY1ZIs4Wa7Z7crkte914kDe997/kmtVrto0SKdrkFwubCwcNeuXYIgnDhx4pprrhFF8cYbbxwzZsw999xTUFBQunl9tkEtYgwAYn3ljPvudrlc4XD4/fffnz179vmzqvBHy30RAkE3q16KbZIIdbXqAUDEsK/ez4kYAWAMe+v8UvKvhibPRjZPnNCnJhV3z1MeOZw4rCfKJcaY3Vc1P9flAocDotFLt5T59VNaWnqRwwkTJkxsrCeo02qH5qQMzDAna5V+htta7S3smN2t/+BYJLzsw/n5+fnPPPNMYt97Hpjee9hIS1Lyj599vGv96pTsHEGlYRvLVUU4IcaLuSbN8GxrrxSjIOKt1Z4IJ/RINozOSbKpFbyIEUDfNNM1HWxGJS3dLCLGkspoqlY5OsfWzabnRLzZ7gmxvDRmTShW7oscdASkeuohVkDQYJx1TzbQCQ9g7SVpQ0ZGpnW0FwOrtLS0s+QGb6SgoKC8vDzabn5NrVbrzJkz44czZsxo8licSKK/CjVXKK3Qpu9s0bKC6ItxZhX98PUjt2/dUldXd9tttzVpGbfq1q1b98MPP0ivGYb5cvUGeyAqjSta0noUD4t32bZtW5NBYry43+HfV+8/6gptrnJLeXcCxiWVrhgv7qnzVQdjOgU1pmOSRU27ouyWKo/0swELFsC6dfDAA83+mXUhZpPdtbvWJ/0knPFHNla6DzkvWwijpXTpAikpsHDh1Rpfpj1x/fXXx1+r1erEJCoAoCjqu+++27Jly3/+859Tp04VDx5EISRlqfsZriIiUARKinrefv21Q4cOpaenJ/Y1dMg32ZJ/+mLxwtf+8t5zTwg+V5AV/rnw01WrVgEAH4t2NKikrHYpvb3AopPid1qa5ES82e7eW+eXDhlB/LnSddgZIBDqZNamapUDM8wEQl2suu5J+lSdUqugAEBDk8OyrEqKOOkNry531oZiJIEwYCVFAECJ3V3hb/ii8zPc2nJnVDz3G6OoCK65BoqK2n6VZWRkrhrtRabB4/H07t078YzJZMIY+3w+tVp9oV6/MLNnzx4/fvyBAweKiori+/KuDF7EnsbwQYgVBErRo0ePi3eJy6UCwKjJUyY9PJPn+dmP3WdOSn7w5Tefe2/x3x6669D2zQDQsGcqARVFDEo3b6/2HnU3bFekCGRU0u4oW1LlVpOkvjGmOSTTurnS7Wc4V5RJ1aqgf/8oL+yo8abVVRYkDFjqDFrVCoOSUlOkPRAFAJtaIQUrk84t6CYjc2XMnTtXrVavWLEiNTX15ZdfzsnJadIAIVRcXJx4hkDIoqalWgVqihwxsF+z2qFFyQazih75+PQbB/To27fv6q5dB4i4g98nWWGqKruxMA9ycwEAtmzJNZ2zA4MVxCgvemIRAOhm05fY3UGW19IUxpBj1OQYzzbubDmn9p9BQQ3NtKw/44pwAkIAkUjHU8eLkg07UzrWiWhPnY8ikE5BSuWl3QjO2Wby8suXtXQyMjLtgdYaWH379p0+ffodd9whSV9ePV555ZVZs2YlnnnppZfuueeeq3dFh8MRjUYp6pwlysvLy8vLA4DKZnf3tAwRUBlDhkSkI7COwHU8lFS6Ois4zUX9iYWFhZ06dTp58iQAxMJhnmPnznhw94YG0YRpf5qlVir1ev0tt9wyZcqUZqeXR6MylgIAAiCP4nTAiSSpEdkUQhAJ5KiplprpMekVie1Vns5Kwbx2jfHVV/oDQlyDOSjccANH0TmAjs38U+ym8XkklAm0PRCV3Gm5NMe66yvdLV0Kj8ejVqtbaENniiIB4PV6g61Y/Ozs7Cvu24b8YjfOrxe1Wj137ty5c+e2vMsJT7jUGSQRBJ31YEv5/tDpiUW559tYCCDLoMZYVVRUtG7duiSvN/HhQQWQ6/XCnj0AAKKY2JEVRE+UHZZlKbF7zvgjVcEoL+I0nWpguumSNWxEDEddIcnXizGYqyt7T74eAIoPHy4xpDoj7J46n1TAKt+itXKhi48mIyPT/mmtgWW1Wh955JGnnnpqypQp06dPv6QI54Uwm81+vz/xjM/nQwiZGot2vfDCC88//3xiA4IgLiSU0CbQNH3FOlgXxx6Ihmp9VrVCkp4qdQaPe0JhlaFLquniHXft2rVgwQK73T5u3Liv5r8et67WfPVZt6y0jat+8Pv9KSkpvIhPeMNdLDppeRhBLPdFulh1J71hcAQAQARQm60dTJpsgPNXMBtA7wmVOoMnOTonyKRVViS+S9bUSGlZ3RRYnZ0NAKIndNgZBIAkjbIo64IiEc2iaeSCLaqqIO6l8PsBwPyPf5g/+ggA4NZb4a23Ludq7Yi2unFk4tSGYoecAYpA7/3p8fUrlv/p3Y97DR355Za9Y/sWpTSWr8EAzjCTrFUCAMMwn3766datW9MBpPSuIQCDANwAnwB06tTpxhtvhHNv/63VXk+U7ZViHJBu2ljplsLoeWYNcakvIhHDtipPfYTRKajeKcZdtd54yhcADMuyHnAETnnDADjLoC5KMtTUyAaWjMyvntYaWKtXr66oqFi0aNEnn3yyYMGCnj17Tp8+ferUqUaj8bLGKSwslArFxDl+/HjHjh3jvo2LJJX/6sgyqBGCVG3DtvDCJH1txel3Z7/Nc9zdd9/tcDhOnz49aNCg86OQJpPpqaeeih8uXrhAkjGjFYobb7s9fn5fvc8eiAUYrn+amRNFKff2YNlJypJCEajAojvmDu2v9xMIEiMaiRRYdLyIj7tDjvwulU//OUuvRm4XzJsHAKfvuIdNTskxqtWDBwHAGX+k1BkkEFKShCPC7Kn19U0ztaXZy/Ngt59zxusFrxcAwN1iR1n7o61unN8e27Zt++tf/+pyucaNG/fnP/9ZoWjprtVUrSrHqLEf3vvTsq8B4I1Hpz36178jUdCmZHSx6rrZ9BhAqqfZO9XY0aiZPHnyypUrAYAkKUHgAeBVgEEA9QBPAcDJk/vuvLPXuQ7sbjbdtmrv/nq/MqEwQ3UwlnypsLg7yjiiDE0SwzPMKhC7G1Wn+EbdFp4PhiI1ngDCCBNEbSgmK7nLyPw2aDOhUVEU16xZs2DBguXLl1MU9fvf/3769OmDBg1qYfdPPvlk2rRpR44ckTYShkKhjIyMe++99x//+EebTO8KaI2S+2WxZcuWESNGSIUOETr7P/Laa681cdo14ccff1y0aBFFUXc+8yKjMXW3aNI+elf//ff1HfK2zJoDAOk6VYjjAwx/aPuWvMIikqLYEwfvvvWm+jCzvdpLEWhsbjLdnHJWlBdKGjdAUQQqzrTYKsuhSxcAWL9stb9b975ppmyDujoY21njRQgNyjAbFFSJ3R3mhLjsdUu4tNBoIAAffNDwetYsiETgxhsbfFpFRTB+fAsv1G5p5Y3T/rksodHKysrCwsJ4uuHMmTPfukwn5ebNm4cNO7vhY/SE3z3xzw85QSyw6mK8INXTNKvoWDj0xL13+l3O6tMnn5r3UW3F6aO7d9y+e8cMv+8IgCRwvG7dulGjRjUZvzbMbK/2SPdoskYZZPkoL+QYNX1SL/aZD7L8pkoPIwj9Vn2bPePRZtuU//M9/v/ukpxwXVQ4Ny2pSX6CjIzMr4s2VnI/derUm2+++ZEUwQEYMWLExx9/3FEq73BROI7r27evVqv94osv9Hr9Y489tnr16gMHDmRkZLTh9C6LX8bAcrvdN9xww/bt289/S6PRBAKBlvjtJINJwHjoW7OSP/p/3u49d323BgMOcwIAHNiy8fVH7unYpTutUNScPOZ2uxFC9WFGRRFGZTN/Xdy6ytSrDUrqiCtIEyi3vqpwWH8AqNiwZV96Lsa4X5opWavcWuXpYtWl6VRSx61V3p4phouUBmrC5Sm522zgdsP8+fCHP7Rw/F8LV3zjtH8uy8BatGhRvHIoAOTm5p46deqyLocxvvPOOz///HMAUKvVS5cuHTL6WilzHADUFDkww7Sj2hflBUEQwsFA9akTXfsOcNfVmpKSkj7+kPrnGy6T5SaXAwAKCgr27NkTr08lYrAHokYl9XOFSwrvZRvUXay6ErsnWavsk2q8REF3li+xu1O++qLvn//YbAPP+x9aHnrguCdU6gzmKHGPzGTZwJKR+VXTNjINsVjss88+GzVqVH5+/ooVK/7yl7+Ul5evX7+eYZipU6e2ZASaptesWZOXl9e7d+/s7GyXy/Xzzz//F62rXwafz9e/f/9mrStIEGi4JCla5YB0MwD4eQAAEqFeyfoo35DkEY2ERUE8cWif0Wrz+/379++XujRrXQGAK8JK1lX/NFMXq67QpudEbAeqbsw4buLEDh3SBqabEEJHXEGKQCM72NIapa45AV+TY4tbV5JUvcxFaP2N8xvjfK2Wyx0BIfSvf/1r69atS5cuPXHixPXXX29Q0kmahs9kul5lUSmk4oAkSRpM5q59BwAGS3IKAvRq+clNny5lN+ySRE+OHz++YsWK+MgHHP49db7NVR4RwKZWIIDKQPSoO3RNB+slrSsA0Cuobja9c9CQ7fMXHPl/n7j/9qZ0/tDzr+yYv8C15AvL9dcBQIFFd00HW6qs5isj8+untU9I+/btW7hw4WeffRYIBMaOHbts2bIbbrhBevDKycn54Ycf0tLSWjhUSkrKv/71r1bO59fF0qVLy8+tKWY0GuPJ/i+99FJL087eftty7HivQNS6dycAaGqrXQ8/WoQxIFT64l8HXXv9jLnvxMKRUZN//++354wfP/7UqVOJdaObkGVQKykiSa2UMncLrDqTiiazrfQPy2klDQBpAIMyzHoFRSbk9rqj7JYqT5pO1S/NhABqQ7EdNT5JKf5yl+WCTJoEgQDk57fZgP892vDG+S0xdOjQF1544Y033uA4rrCwcP78+VcwCEJo8ODBACBg7IlxZ3yR6mBMTZE8xqe8YQWButr0Q7Ms6844Ja03QIAQwdTZH5/1N4FSnTp84NThAwDQ75prtYaz+07yTNoz/ggriFqaRAgwAEJgD0RzTZq4qvBFcEbYA46AkJ7JZWbViJiqO2MFAID6wcNSB/SxJZ29Tcwqur2o/8nIyLSC1hpYffr0SU9Pf/TRRx944IEOHZpWCLZarcOHD2/lJX7DcNw5Pp78/PzS0tKVK1eeOHEiLy8PY3zgwIGePXteeqBvvlFu2pTbeES5nLn/WiS9PvCX1yiEisdNBABPfd3mFf+pr6/fv3//+VpZiTRJ2o1vwoqTet4ZNUUqScIeiGIMGXrVrlofxth0ASfZFXJuRblfNW1447S3KulXhtPpnDlz5tatWzt27LhmzZqcnJzs7Owr2yaMMV68ePHevXuLp9yrsqWIGNQUWZxp2V3r84viUXdIQRKOCCsmJEcgAFVqlgAAgA9u2RgOBIb/bvLjb8zjSbI6FJPKERqU1IA08+46X5gTwpxgVNJdbTp3lLW2ICDuirJbqz2CiHsmG5K0yhK7uzIQjTvrTvkiKVpV3M0mIyPz26C1BtayZcsmTpx4kVyBtWvXtvISv2G80m64RqSUrxtvvPGrr76aMmWKVFLwsccee/vtty8+jiOvMxmMkAhpqisV9fW8Rusv6KpVUIKIuXAQtHoA8Lkcs+79fW1FuVKp7NjpnEBMZWXlhg0bMjIyRo0a1fK4ZBM0NDk821pi91QFo1XBKALok2rqYGwvIrHtjba6caQq6Wq1eu/evVLy4siRIw8ePPirc4BNnTp1zZo1AHD69OkDBw6cOnUq0briOG7Dhg3BYHD06NFx6ZZmYRimR48eZWVlAHDCHbz/hb8SBCqw6vbV+/0Mh6NhmiJrQ0pHhFFTZIwXGoSpzg6Abnrw8Rsm30pbUwEhAeO6RgMLAFJ1SrOSlnb5WdV0uk6V3vjWheBETBNIQ5Eqkuhk02YbNTSButn0R6y2sumPpepUnTt33C3irdWe4gyLbGO1QyZPnvztt98mJisrlcpXXnmlSf0lGZnzaW0O1k033SRnYracefPm9erVq3///p988gkAKJXn+IGSk5OlF0888US8YPP8+fObVGHjOO7TTz998cUXf/rpJ+mM4r33Dv3nxwPfrKi4fhIAhHI7bfxi+dol3/605NtTx49JbVy11dXlpwBgxKRbt3uYcn9YwBgAVqxYUVBQ8Npb/xw3bvzEiROlzYxNCLI8Pvew2b9OTZGdLQ2RR7Nakd2cdRXjzxmfFc9v8j9BW9047b9KeksQBGHdunXxQ5fLtXfv3vhhOBweMmTIddddd8stt+Tn5x86dOgiQ7311luSdQUAqz7/5IMXnwYM++v9nih74uC+/xvW5+6RA0+cqVATwPACABTadGdVrBo/5bQtDRACgCyDuk+jOp2A8dYqryvKGpW0kiRO+yIH6gMXuhckWEHcWOnaV+/X0OTonKR0vWr9GedRd4gmCM6WrJrzpnbum1mdOvROMQIGXhR5Ua5D2L4wm83Lli0jCOKuu+4qKyv76quvsrOzGYZ59tlnr/alVSrVVRV6lPkFaC+1CH9dCILQrCHSLCzL2u12QRAWLVo0Y8aMAwcO7N69e9q0aStWrLj77rtzcxsie1qtVrppRVH0+XyJIyQ6ujDGN9xww9133/3qq6+OGzfu1VdfBQCTih6QZg41ZrXTBEEgYEUcjYS79Okf8vsAcKei3q8s/nrULbdPf+kNUcQH6gPbqrwCxs8999yAsRP++vnyGXPfWbly5fmOE0+M21Dh2lvXUHCwPsysP+NqtuBgXYjZXx9AACqK8ETZ3TW+Jj8XR13BNeUub2ONoAOOwB4/F5+2zBXQ/quktwSSJBPDmkqlskuXLvHDTz/9dNeuXdJrl8vVpKJDE44fP554uHnFdzQ0fMD+s+C9aCjoczufnDz+jsE9P33rr0fXfu+J8SLGlWXHln/8wZfvvJV4X5tVdL8EUbf99X5nhEEICm26oZlWEqFTvvD6M64AwwsYH3EFhfN2ZDOCyApiuS+yv97PCuKmSneYE4IMn6FXXZeblKlX7aj27qz15hg1Y3OTbBrl5ir33jq/bGS1E6677jqfz6dSqXieX7x4cX5+/q233lpRUVFWVtZ+CrjJtGdkA+vy4DjuwQcf1Gg0Op1uxowZongJ4+Df//53SkpKdnZ2t6KixBR+rcGwcuVKm822d+/eBQsWzJs379ChQwMHDgQAgiAeeuiheMvevXtLhRpTUlLuuuuun3/+efXq1fF333zzzfgcSAQqmhJpOkaSIgYEoNZoPfV1f7ptwqK/vQwA3foPfvTVuYIguhx1IgZHhNlW5R1z5wNPvPk2RVO1FWfg3IqHEiqSUJJEhT+6q8ZX16gHoaWbel8cEWZbtUfEuHeKaWS2TUuT9mB0bbmzya8FJ4qb7W5PlGuQrsbnhGdkLpf2XyW9hXz++efdunUDAJvNtmTJkkRZhyY1HpocNuHGG2+Mv1ZptHO/XsECoaVJjPETs+cXDR4KALFoxOd2fvvRu8//4cEdK5Yd27vzL3dOWvzmKzVnThIIxT+R3hh3yhuOj9bVqjeoKIxhZ63vhDckmVM6mqRJtK3ae8zdUM8gEb2CGpZlVVHEaV9k7RlnmBMy9er+6SYEoKZIVsAhTqgOxnbVekmEttjdnigXZHlR9mO1D6SY9ZEjR5qcz8/Pj0Qi8cNp06ZJZUUkEquSKxQKhNDXX38db9AkwD1o0KB4R4IgHn/8cek8TdMMwwBA/N2CggKE0JQpUxK7S28lXuu9996LX6uJavFF5ilzlZANrMtj3rx5UhQmFovNmzcvLlzULE6nc9q0aT6fz5aWPuPtRXnFDaKFGbmd5q3YOGD8jQBgNBrvu+8+q9VaXFxMUdTw4cPPnDnz1ltvLVy48N57733llVcWLVp022237d+/3+FwLFmypElpNoqipBtMQ5PFqQZy1gsbymo2frEcANK1qtoTR1+8a3JtRfmafy8+ULIeQLojwWS2IASAwBFhht54CyBUUXbsq3ff6tKly/jx45lzXUoamhyWZdXQZFUwurXKI2DcO8XYpAIuAGgpUppJfSSmpAi9kgIATsRcQkmQrjZ9V6uOE/FGu+uUN6yiiJ5GSkfLH8Irx+PxNPkajVdJ/29N6cro3r17aWmp1+utq6u79dZbE9+aOnWqxWKRXhME8YfzVND27NnzwQcfSEHGm2++eeHChd27d8/NzX1/5Ya0vM4WtWJUTlKw7CCtVD73/uJpz8365/frU7NzACAlq0OlP2YwWxVKdeeefWfMeQcQAkDbVv1QfuQQABx0BCoDDabq/2fvvOOiuLo+fma2s8suyy69Y+8FUFHEikmMnSQ+lhSTGJ9YY/SNsaVZY4oxMWo0tuiTGCN2o2JXLBgsKAqidFhg+7J9Z2fm/WNwXEE6KJr5fvhj7+wtZ5fZmd/ce+45bhzW4BCvbj4SJ0HSBxEEuVZiUJrtEh6nnUwElXDnsiP9PADASZBCDivKZUqMz0b7BnmKuOxCo+1YtlJrw6j0Wawnxf5leMpgGEaSJIIg1Yeje/PNNyl/j8WLFx84cEAgEJw9e/a1115zrTN27Nj+/fufPHlSKBQaDIZXHoZH7tWrV3JyMpvN3rRp07x580iS/Omnn1atWkWNTjmQkA+5ePEiAOzZs4fudsWKFQAgk8lcx5o+ffrAgQOPHz/u5eVVVlZGu53Uxk6GRqeRA42+SDwx0Oi4ceN27dpFFydPnrxx48aqekhOTqZCcoe17/jF1t0iicft00c3r171xbbdHjKvMImgm68HACxbtmzRokV0q5iYmAsXLtDFnTt3vvnmm3TR39+/b9++f/75JwAgCPLjjz/S9xuTxXpRoTfjIHfjmhxOm5MI93DTpKUUFBR0jBlQgHOpuS4EQQBB2CiCE/T/nrxzNME/vPWYvj3yHGixyR4bLBNyHtt5nmewXCsxAICngNsvWPbEO4AVwy8Uak0OJ5+N2pyEiMvuG+QpqLSD/WyeWmvDACDKz0PgMNch0ChDJby9vQcNGvTHH3/QR1auXDl//nyFQkH5ue/cuXPHjh2uTd58880hQ4Y8NQsxDNPpdPSFvn4UFBTs2LHDYrEMGzasQpj79evXf/nll9R1bMKECd9//z39ltpBFtvJdiKUjQAAnE29q3Ui+TlZkYNeUSkKv5k5+cvtfwmEInOZYdGE0SibtXL3ETabcyXxyPcff8hisf9OyzES4M9DwlzSsBMANwy4lQAAYCOIkyQBQMhC2orQLDMR7oa6PX6+2wnytpGwEYAgQJLgz0fDBY/9eiw4eaOMIAHYCBIpRqj0WSUlJXK5vPm4tzbw3/c8curUqcGDB7NYLKezOjc76qlSpVLJ5XL6CIIg1MWWy+ViGNalSxcq+iD1LoqidN4OAKBvwXv37o2Pj6dH5PP5drvd9QZNHcnMzGzVqhVdTEpKoraEU2O1bds2PT2dqs9isQiCoGyr3k6GJoIRWFXyRIG1cePGKVOm0MXff/993LjyJIA4jtvtdletYLPZ2rRpk5+fDwDh7Tstil6u0AAAIABJREFU3bmX5yZEESBIaCEVUgGiCgoKQkNDXc9yHo9ns9no4p07dzp37kxXGDFixL59+44ePZqVldW7d+/IyEi65qUCdYkF83bjRQdKLRh+oUBjcxIRvh4WiyXdYAeALp58Dl9wrVgPSKX/O4IASbY/fhC9dcvSrl34jP+KuY8u7kqz/XKRDidJLgt14ESIRNDd98kJB61OPDFbhZMkiiBxYV4VVBoAUCuDbBRxEiSHhXYSod5iISOw6k2bNm1atGhB5dSj+PTTT1etWmU2myk3kaysrAcPHrg2admy5dMM4etwOHQ6nY+PT1N0jmGYr6+v63pNWlpaixYtqmmyZevWdIN9wOixNouZ7yYEAJNBv+mrBe8tWiqWep7Zt3vdojkEjo8aNeqPP/4wOnB37qNzGCfJayqz2orxWajNZWrWncMCII0YESDkdpE/ijBnw4nkUpMZw33dOK0kgqtKkx0nwsT8dtJyDx4nQf6jNOnsTmph0s+N29VLiAAUFxd7eTWjVDl8fg2bJV88Kgusq1evUl4cFJs2bXrppZeCg4M5HA69JwkeF0aU6HGVNSiKUjNS9+/fb926NZfLpZYCKVwlV2WBtWLFigULFgQGBhYUFFgsFqFQSGs1eqz8/PygoCDqSJs2bTIzM+fNmzdt2rTq7WRoIprLD/h5YfLkyaWlpZs3b2axWB9++CGtrtasWfPZZ5+VlZUNHDjw999/p24nfD7/xIkTn3766f3793v0iOrqI0434gQJUj6HDr+Zn59f4RmCCoCE4/idO3ekUmmHDh02b978ySefqNXqQYMGbdiwAUXRV1991bUJhmEZGRl+Uk+Sy+oVKGUhCOX8kaUza7IyZi5YPOObtX/9vHrm4YRLly5F+nunKFzCQ1CXdpIEAPejRwKOHXKOHMWe82gVRmVxUOqqm4/EW8i7UKDJM1hRBOlWKeEgScJtZbmrL0GSaaqyHn5S130wlLris9G+QbKCMmuGxnTbSPYQ4Iy8qjc1Zklv0aJF9YKjqUFRlMfj1fIObXMSZXbM+2GUNTOG25w4FWiKIIh//vnHZDL16dPHtbcKweSgJjXwweTJEyZOzO/YJbhVWwCQcRGQeMz+bh0ASAmb5f6tbl27WiwWrVa7bt26WbNmuQb7vVioVVsxDx4nOlCaVKClthAiAEYMBwAvIS8yQOoafZd04ghiDnDnUz+EWD7vQoEWUJSy0EmQV4u0OrtTJuB295VcLtIVWxyoFu3hJ6W+seYjsP6FUNdh14tzjx49KDkiEonMZjMAUKvSGIZVv92PVleunDt3DgAqOMsjlR99XZg/f/7ChQsLCwsBgEq4OWzYsAp1aHUFAMHBwZmZmZcvX6a2jNRoJ0Ojw7i/1A0EQRYvXpybm5uVlTV37lzqYHJy8uzZs8vKygDg9OnTtKMiALRu3Xrv3r23b9/+Yf0vOVYSAFAEdDbsrrrcHzYyMtL1/ufv779jx47MzMw2bdp06dIlJCRkypQpb7/9tlKptNvtJ06cqBzf6ObNmy1atOjcuXOr0JD/ffPVwf37U1JSAMCdy/YhLLGxfdXKUg6HG//hLGlg6Joff8zVW6hfcLmrR/nP+dGvmv24C4iYxxZyWV19JGEebkIOKzZIJuKyK2cbJAFSSvSFRqsbh9U/RO7OZRcZbVeLda6XCzGXzWejMYEyKm1IO5kIBZLN/OYbwIgRI7KysuhFAZPJdPr06REjRjxbq+oHQZJJBZpLRbpikw0AzBh+Pl9zsVBrdDitVuvAgQN79eo1ePDgNm3a0AkK+Xz+Rx99RPcwbNiwGoOsstnsxT9tCm7VljrnzQRKSSIWgkSEB3311VcPHjxIT08/f/78nDlzvvjiC9e2IRKBlM+JCfLMM1iNDiefzXJjs+gTXPx4bgMAELBZ/YJkPfzLHzPcueyBIXL6yYQgSQwnKb8rdy67b5CnkMMiSCCZfR/NAA6HQ8mdCsk2XKHW5gQCAVmJGvvv168fALguVoDLcmFVUIuDCxcuvHHjBgBU3i9cUFBAv6YWT6KjoxtiJ0NDYJYIq6T2yZ5/+eUX131/rVq1oiPxUBgdzgsFWpsTD/NwC5O4JRVqHThBeXwDQH5+/hdffJGVlRUdHb148WI3N7e4uLhTp06xOZxB8eMS/9wxfPhwFEUHDBo8bMLb4dKKjrTR0dGVsxm+9dZb27dvX758+cKFCwHg1Tffm7TgK4fdZtGUSgNCAUDC4/QOkJ7KUztwwvfcKbbFjGFYu9+3ed68hkdFsSjtKBTCq68CAEGSqMudo0KRAiOICwVaJ0FSfldUxmg2ivQNknFcFBsVd5EulqjUYmG5D5bT6Tx+/LhSqRwwYEBoaGiNXzsDNMss6RWoZbLnK1eufPPNN96tO8a9PQVF0U5e7plas9WJh0jcuvtKftmw4cMPP6QrT5w40dWxbP/+/VevXm3duvWECROoHywJUJVsv6M23tOYeCy0T6D0RqFahyMAIOayyxxOHgtV/nP23fGPNmpV/i1TPRMkeUdtDJe4JRfr9TYMAFCAPoEyL2HdIoU6cAJFEPqRxuYkuCwERRCFQuHtXSnZs9MJ1A7iqCjw8qrTQAz14KWXXkpMTBQIBK5r0PBwBmvTpk3vv/8+NSek1+sr7DWhoJbtXG+y9BIhVPLBOnjw4MiRI+lFSaFQaLFYKtyg1Wq1l5cXiqIEQXh6emo0mgpjtWvXjt72WNkHqyo7GZqKypKWgUKhUDgcjtrUTE9Pd9VhU6dOrVDBaMf+flB6vURPFfU2x6H7Jelq4xO7oiMAffDFyoQMxYdLvkVQlCdw++q3PQkZiiyduUKTqvxPL168SG1IoRg68d2EDEVChuJAZnGG2mjF8MRsZUKGYl96oSkwmIqWUPEvKKhO35jDiVswJ120Yk67E6++iUqlMpvNJEk6HI7+/ftTpvL5/KNHj9Zp6H8zJSUlEyZM8PDwoKT57du3n7VFj2G324uLi6uv8+DBA9oPL+bVkXvSC6lzNaVYT5AkSZIVZpKGDBlSTW8ai+NEjtLkcJIk6cCJIqP1VI7KiuEkSdqd+G2V4fD9Eo3Zeurc+YS0vIQMxcHMYpPDmVpqSMhQ7LubL5J41GYguxM/maNKyFAkFWjuqsqofrTWWl0xaqSoqIi6MT+GwVD+wzxypFFGYagRKqoCi8V67733srOzjx8/PmjQIOrc2Lp1K0mSEydOBAAEQaZNm3bnzp1Zs2aJxWI2m001L5f7LtCiiiTJqKgoAGCz2Vu3bl28eDHV7fLly6l3qfArffr0oa6QNHR46pMnT7oep8ZCEGTw4MEnT5708vICAJlMRr1bvZ0MTQQjsKqk9gKLJMnDhw9HRkYGBQVNmzatwu+BwoY9JjWs2JOVh2sGOt/g0I1nryVkKKYtX02pq+/2n3BUkiyVd61THD58WK1Wt2zZkvpdTfliJXXT2p2Wl56vSC7SJWQoLhdqbxaqjP4BjSKw6gEtsPbu3etqfGRkZFMPzfB0qI3AWrNmDf2v9w4M3p58hzpXFUYrVSEzM9M1Pfnu3bur6e16iT4hQ3Esq9Rgw07mKPdmKBIyFPkGi82Jn8hRJhVo8kuV1ErimA9mzP12bVGpimqYWmq4pzHOmjWLShgVFhaWnp5e1SgFZZaEDMXJHJWTIEiSTFcbEzIUSYWaun07VcAIrObDqFGjKrgusVisESNG0BVmzJjhWoHD4fz888/UW9ULLPKhxqJAEOS///0v/ZbZbHZNXEYfp4JL0zNhNNRYrrto3d3dXStUYydDE8EsEVZJ7ZcIq8FoNGIYRkfxqRFfX9/S0lK6GDsobtz8Jd6BwQCQczfth9kf/O+37RXSADscjlWrVq1bt664uJg+2KZNm6tXr6rVarFYfPToUSSghVtAmFZZenb/7qUF+e737nLGjyt477/tZO46m+OXn386dDSxXVSvV77+6nWAG2FhAbfuePA4XDYLmthBSq1WU2Eafv/99wkTJtDHO3TokJaW1qRDMzwdarNEeOTIEcpd1zsweMmOBLlfgFzA1VgdCIL0CpBSmcVv3769bt06i8UyduzYoUOHVtMbCXCtWJ9fZqWzDQo57Jggz8tF2jK700fIO7Lh+xXLltH1FyxYsOxh0UmQBjvm0GvUanW7du0IBDVjuJT/hIsAQcLxHKUNw6mcm2V257l8NU7CgBCZpME5zh9bIsQwoHI5mExA+Wv+/jtQ8ygiETA7cP9lxMTEXLx48ZVXXnHdOwxPWo5keOYwTu71xOl0rl69evTo0R999JGrsqHBcXzKlClSqVQulw8fPrxyhPQn4hqHWiKR7Nq5g88u/x/l3rtTVJDfv3//33//3bUJl8sdO3asqw08Hu+LL76gto+1atVK5OUrDAjTKks/e3PM6YRdont3ZbduaO/e6yB311gdSYXa8P7DZn63LnLQyzy+AAAIhHVJYUhS6Mkq/VganxEjRtARyREEmTNnzlMbmkFhstGRMwEgz2Ch3MyfGkOHDp06dSqXx/t8yy65X4AnivcNlnX3lZAkeaVIR23W69Sp0/r167dv3169ugIABCDCz8NHyLM6cRJAwGGZMWdijpJSV70CpMqSEtf6SqWSekECXCrSJhVoUbFnp06dbARcyNdeKNAY7BgAOPDHdvuiCHT1FiMIcr1En64xXSjQYAQZ5uHWcHVVkQsXwMcHfHyA3g0zfnz5kcfDDjO88Ozdu5eKOFpBXTE0T5htwPVk5syZ69evp14fOXLk9u3bZWVlhw4d4nA4I0eOlEgk27dvp2OQHj58eOnSpStXrqyx2zVr1kil0sTERH9//8+/+PIBxhb7Btr0GosdGzB6LF8o/G7WlNWrV48fPx7DCRtOuHPZUCm/DUmSM2bMUKvVAKDX6996fUxy+oMx094pzstBWSwTkHIABJA8g/Wm0oATpETkdjrxWPRLwzr0iIbzp/3CW+WTZAsPt7rOXuUaLMFiAeX/TgLkGSwhErda9iESia5cubJly5bS0tKhQ4fS/lgMTY0DJ1KK9ThBkiSESAS5BsuNEgMLRV4O9+ayntIDGIIgP//88+LFi/N1RrZE0M3XAwEIkbgBgMmBu3NruEzl5uYqFIru3bvTARpUZrva8jDkD1m++52NIpF+HiwEmTRp0vbt2ylXYjab/c4775SbARAsFmgsjstFui7e4luqMpwgvYQ8EZetstivKvR+In43XwkCoLU6LE480F0QHSC9UqRLVxsBIMzDjQ6/wsDQ6NALfLTDFkMzh1kirJLqlwjd3Nxc071t2bLl//7v/6g9HcHBwZcvX/7++++/++47usKwYcMOHTpUJwOulxhyDRYPPqf0yun/m7/w+wMn2VzugS0b8q+eTzx56kKh1urE+wXJRFw2SZLR0dHJyclUw8mTJ1fI4ZOens6bN+9UYqLNZnsDRb0JwtC2vTqiJwDgYeFpE9/j2E337tzpJPfh6nWYh1TWqX2Eb902m2RqzWmqMj8Rv6e/B4Ig1BpNa09RRy/3alrRS4R1GouhcSk22ZIVepIkSzPTvFt1oBbm/ES1ilyl0+lEIlE1K+n0EqHFYvnxxx/T0tK6du06ffr0xopdOWPGDMqVxMfH58iRIxEREXRc3C7eYo3VUWi0AQAV25aKicBGkUuXLm3ZssVsNs+cOTM6Otq1Q0pi0pdFf3d+B7n7qVwVQYIHjzMgVK6zOpIKtThBDgr1QhA4n6+x4wQARPh6hEjqkwPY5iToiWqqqFWWPFoi1Ongxg0AAIsFhg8HAFixAnr0AAAIDwdmvy0DQ3OFEVhVUr3AatmyJR2MBwDi4+MTEhLo4qJFi/r37x8XF0d/vevWrXPdZ14bbE78tsrY1VtMOLHBgwe3iBk84p0PrGaTr9OI+ATrbZiXG7d3gCcLRW7evBkTE0PFvhOJRNevX3/rrbfo2A0dOnRITU1l8fnwpJwPzl7RZ/84UGZ3SngcaimEhcCQcO/KKW6qx+4kLhRoyhxOPxGfgyL5ZVYqiWHlYO6uMAKrmbDv5FnMN5zN4TgxbPMXn+xY92ON2VGysrLeeOON69evi8Xir7/+2jVSiSu0wBoyZAiVPRcAhg4deuTIEbqOGcPp84QEsGK4W7WnDcW9e/dOnTo1bdo0+kh0dPSlS5duq8rua81dfSSB7vwLBVqDHQMEIn09svUWrdUhf5jvjyTJwsJC18CM5QbjxJk8tRnDAYDPZtmcOIogBEkigJBAtpAK8wwWJ0G2l7sHuPOpfAl+In6p2U6SZISfR7C4bhpLY3VcLNR2kLu3kAoBoMRsT1bowrhE+4BKYRrKyoDaY3/kCFRYJ23ZEnJzYfly+OSTOo3OwMDQdDA+WPVk/fr1YrEYABAEWbhwYQWJ4HA4Bg0a9L///a9Hjx6dO3f+9ttvW7VqFR0dHRwcPHnyZCokaY3w2awoPw8OC+XxeKdPnx4b28NR8EAgFBkkfnobJhdwowPK88KuWbOGUlcAYDKZfvnllwMHDrz33nvdu3efOHHisWPHWCwW9O5N/ZEidwCwefkYIqI03SKLw1tH+Xm4cViUuuKzUZyECwVaqxOv0xfCY6N9g2ViHrvYZKulumJoPhw9fpzassRis0kWxzW5YVVMnjz5+vXrAFBWVjZt2jQ6+s4Tyc7OptUVAPz999+XLl2iXqepyk7lqv65e+/dd98dPmLErqRridnKHEP5+WzB8NN5ap3tsXDtBEGMHz++bdu2ruoKAKgg1528xP2CZeEebmwUceOwfIS82EDPYLEgJtBT7sbls1E6ilvlh0sHTiQVas0YLqayleMEABAkyWez+gRJ2SiSpTNT6qq1p+hSodbmJFpIhdEBUnritsxeXeq6ylCPYFSSgxKz/UqRjiDIOrs/Op2A48DklWNgaE4wPlj1JC4uLj8//8aNG6GhoaGhoSkpKQkJCVQ8Ok9Pzw8++AAAxo0bR+XSyc/Pb9u2LbWk+Ouvv9pstgr5d2uEw+G88sorGEGczFFT0qed3J2OT1hhmo3D4Xh7e//666+PdXHuHAAYHU5nz17Sm9dsY8ZINqxLKdYXlFkFhVqrs/xeI+FxZAIoMtovFmoHhXrV6TLPZaESHoe6wbhz2QI2I9+fD/LLrC+/N8PpdB7YvH74pClTvlhpeXC7xla3bz+qQxDEnTt32rdvX1VlDw8PFotF500DgI0bN/bu3RsAeGyWkyCznNyUjPu9Xx7O8/InAG4pjSIOR8hhXSjQmDE812CR8h+tWR88ePCJEpBOi05l10ERpKe/BwlABVgvMtosGB7h60Hpq/s6c7aN7YsTtKsZCXCxUKu3Yd5CXnSANFtvvq00Uv3YnPh9jQlBqMRSwEERFIEIP48Sk51aBKcWx8scTkqZ1R65G7dXgPRyoS5VWUZ1393Xg2PWPaEqnw9ffw0A8DBUHgMDQ3OGuQXWH4lE0r9/fyrmeGRk5M2bN5csWbJy5Uoqd41rzYsXL7o6bLk+ytcejCCSCrRWZ/liypUiLf1YP2fOHJlMRr0OCAiYOnVqVZ24c9k8FgoAHjwOAhDp5+En4jlJEoCUC7hiLrvUbCdIJEjM7yB3r5O6IgGul+gLyqxuHJY7l11qticr9ASz/tzsceDEzVIDgiDr5s/a+f3yb2dNJghC0qZLhU1zlXn55Zfp1yKRqE9MjML4aO+hkyBdtyJ6enpSuUFo6ByCraRC44M0Lo//+eZdQ8a+qSkpPrN7B06Ql4q05/I1ZgwPEgu6Pp74sqioyLUok8mGDh36008/USGCXEERhE5fo7TYLRh+sVBrdeL3teY0ldFOgsnxaMKJIMgWUqGvkBcdIHXgRJbOAgBubFZcqJeAwyq1ODCcCJEIWAiSqizL1lvkAq6ri6GfiN/Gs2Kihdrg7cZr7SkEAJKEILGgSkcuLhc++QQ++QTCwwEADAbo1av8j9pEvG5deXHBgnqYwcDA0LgwM1gNIi8v75tvvsnPz4+Li5s6deqiRYt27tz58ssv6/X6+Pj4VatWUZ68nTt3ppIbUK26detWp1GMRuM333wb0Pclr+AwKReNDfG6rzPfVRuTCrRiTd6aVSu1Wu2cOXPEYrHVau3Xr1/1yRDchALgcoHNBgAEAAEEw8lwD7euPhLKj6rYZOvhX1sfZ5pMjSnPUL4yyEaQ8wWaYpPtjrqskxezr6pZw2WhvQM9MZyI3rT+xIkxHh4eXQOkbjxujVsI161bJxKJTp065e/vv3TpUgUI8hW6jl7urT1FToK8XKRVWxw9/aVyXnk/v/zyS0REBLU+zuPxZsyYQXclR7Dbqddad4kAgN9/WMm3GadOnZquNloJ3FvIIwvvT/1qCwBMmjSpR48eADBy5MhFixbp9Xqq+cKFC2fPnl3ZQo3VwXXai4uLw8LCOBxOuERIklBotJ7KVTtwgo0iLTlOz4dZNXGSpFzj+wR6shCEBNKOEzw2Ghfm9XDeCqR8ToSvR5DYfrlQl1pq8HLj1rjDsTYUm2z3tGYEAEWQ/DKrTMDl1aYZhsHDfS3lFBQAlYouMLDhVjEwMDQQxsm9SmoMNGo0Gtu3b0+5fQDAnDlzXnvttT59+tBC6pNPPvmamtIH+PXXXz/55BOdTte7d+9du3ZVdq2thsGDB586dWrw6+NjR8Sf37HxxPFjAJCpNeWrtKMiO5pN5Xmjp8+YuTdhj0Kh4PP5X3/9tWvO6WrACDLPYGkpLQ+TbXcSRSZbuEedvc4xgrxeou/oJaYm2OxO4qbS0NVHwqv2Ps04ub8wUM7aToJsJ3dXW+wqi8ODz+kb6EniTjrQaG5u7s6dO+12+3/+8x86KzMJcKPEkGuwmMoMIrHEbrW04BM6roRyM0dIcsnkCTcvnY97fcL5A3uWLPnKYDC0ads2tGuP7T+sUiqVo0ePpuMsuJKuMaWrjTtWfbV/y4aAgIBNR07b+O5R/tJMrYnKHhjpK0HKNPQvESPIiwUarQ2TC7hR/h6Xi3TUPpLoAE82iuhsWKnJ3lZePkFFTYaFShrhvC0x2a8odCRJdvf1ELBRSuSF88mOlZ3cK2A2w8KF5a+3bAGjEfr0gchIAICuXeFJ3wkDA8PThBFYVVKjwNq/f//o0aPpolwunzt37qeffkofiYqKunr1Kl0kCMJisYhEdVtEKCoqCnz4PIqgKEkQOTk5oaGhdpzYd+Petu9WHt/1GwB4+Qd+vvXP375ZcvXkMQBAUfTevXtUnpwGck9r8nbj0cGs76qNQWJBozy4AyOwXiy0VkdSodZJkADgwePEBHlyWWiNkdxvlBpy9BYBmxWGWAoMFqOblDoeJBZI+ZzUUoPDZk1OPBo7Mv7S0YM/zJ2OoMjcNZuiBg5pL3dvK6vy13Qnt/CWwcHl8bd//WVAeKu4Nybw2awQieCepjxonIjLbsGytQh+9KiDEeTFQo3WilHbBml11TjfThXobVhSobaTl5haGVSa7ckKXTiPaONfk8ByJTQU8vJgxQpwuf4wMDA8W5glwvoTEBDgWgwMDIyIiHA9Ekk9TT4ERdFq1JXVibtGRqCLHh4eQqGQ2iRIEoRAIKAS7xhsGFckmfzZcsxhT7t6mUowEtk/jhJYBEFkZWU1XGCpLY47KmMmao4J8pTyOTdLDdl6S7HJPihU3sCeGV48JDyOmMfRWh0A4Cfi1TJOqVzALTXbYwI9RVzvtgApJYZ8g8XfnR/p54ET5OHDh4Oi+lot5uK8nN6vjAAE4XB5UQOHFGU/aOGUg6xdhd7OnDmTlpbWpUsXApDbd3M69uz9zqdfAIBBp1GZzbbAYAJ39gmWFxrtBWXWeyjbzyUkBAdFevhJE3NUBEmyUKSHv7Rx1ZXNZktKShKLxVFRUXTQSA8+56VwL87DrHPeZdpXCtL1BgP4D2/EoRkYGJ4+jJN7/YmKiqI9Pzw9PdeuXTt48OBVq1aJxWIURUePHl2b0O0URUbb8WwVna4k12BJzFZRPsJCofDnn3/mcrkAwOFwfvrpJ7FYrNFo8tNvd/HkkyQ5del3K/88LPcLyLp2ZcPn5VFwfH19qS1aDUTuxm0nd8cIIqlAc1Why9ZbeGy0h79Hw3tmeMGg3NK1VoeYx2ajSLrGdF9r1lgdqWozPUteYranKivGKAkSC+JCvUQP50QjfSV9Az17+kudBJlUoAlq2eaLd97Y9NX8VdPfwxz23i8Pjxo4RJGT9cU7rwsqBQH5+OOPBw4cOHPmzH79+qWVaCMHxNHr9UCSXoHBuBO7fv7MzkOJrQRooDvfSUJKif7WQ5MwgkxW6AiSZCEITpDJCp2z8bZp3L9/v3Xr1nFxcT179hw8eLDN9mgHAMclpy8cOcIeNEg2cWLdeo+Kgj59oC6OBwwMDE0Ns0RYJdQSYVFRkVAo9PLyqqpabm5ufn5+9+7dXWenHA4HJYlqicJou1qsI0mI8PNQmu35ZVYUQXr5S31FvHS1ESfBB2zp6elt2rTx8fHZtGnT9OnTHQ6HUCjccfg46RcOADIBt4MQPv/ss2vXroWEhHz55ZdtG28vd4bGdFdtBAAeG40NkjXW+iAwS4QvEFcVukKjjfK7KnM4KX8sIYdlxnAZB2LD/Eot9itFOpIk+wbL5IKafx1mDD+fr7E68VN7/vhjzaovtv8VGF4+I5t84qj6xqXNmx8LRFJaWurv708rKjeh8I9/0p0omyBwBEERBCnKfrDxy/nTln/vHRB06/ypuRPiryj0ZTi4c9n9Q+QAkFSg0dkwLzdulJ/H5SKdziWWb8O/n3Hjxu3atYsubtq06f33339CvV9/hcmTSTc33GCowxIhAwND84OZwaoStVodExMTFhbm6+v7/vvvE1UE8QsNDY2Nja2w9lcbdeV0OlesWBEdHT1s2LDs1JQeflIEgZRiPTWP5ePG9RXx7qqN6RpTjt4iknrGxsb6+PhoNJoZM2Y4HA4AcJNIVVC+30hrdehJzmeffXblypU///yzEdUVANgJrOCnAAAgAElEQVQf7tgnCLIRn+kZXiTayEQ+Ql7fQE8OC6Uy0viL+H2CPIUclgaDCwUaSl1185VUVlf3798/ceIElWmKRshhxQbLuAgMem3c2uNJlLoya9W41dwz7pUpS76tcCI6nU7Xx8XXp33sRNkogqAoC0EQgiD8w1rIfP0+ezNeWVTQOXZQYo6SUld9g2QcFGEhwGOjMgE3OsCTz2bFBMk8+RwOC0XqmpKzCuiU0k8sgt0ORiMYjWC3lx+hikYjEz6UgeE5hRFYVbJixQrKRZ0giM2bN//555+N0m1mZub06dMnTJgwduzYBQsWXLly5ciRI0OGDLEqi0IebkpioWix2X46T52hMXFQNCbIk3bPKikpsdvtACD3C/hy+19yvwBvHqt3gCeKILc1ZqXj0Q0mJSWlW7duCIJ07dr1n3/+qbfBVIxpHhsN93DDCDKpQKt/PKw2AwMASHicPoGenId+VzIBt1eAVMRh9/YT81FQWx0ESXb1kVTeeTd//vzWrVsPGTIkODj48OHDrm8d3LP7p09mAADfTQgAmLksPqLdsI5hIi670GjN0DyW4zwgIOCtt96ii958ljuXLeKW/3D0amXB/Xu3rySpFIVfThrrxDBgc1kI9A3ypPIAogjS019KZSoEAA6K9AmS9fCTJh4/1qZNm5Hv/nf64iW5ublUb3fVxiKjDQDSNSYqBQIAkABpqjLXwFquuM5Xubu7jx079rG3Z80CsRjEYpg+HQAQi4Xt6Vl+JDOz+m+egYFi586dKIoiD6HiXTM8Q5gp6CrJyclxLd6/f7/hfSoUiujoaK1WW+G41Wo9f/uepJ0bFYCHIAgUQfQ2DEGA8i6na7Zr165bt243btyQ+/lLZPL7/1waPTEeAejh75Gs0FvwcoHlcDhGjRpFxWNMTU0dNWpUTk5OnVYtKdRWB6WuqJVBHgtN15hulBoGhDBO7gy1osyBOx5OwSgt9lCJm+uUUHp6Ou2qaLFYpk6dOmzYMKp47dq1abM++mzLLgDAcSeLxQ739XHjsgGgb5BnmspIxxah2bJly9ChQy9evBgZGfnGuPEXC3V6O+Yt5KEA4O1r1OudDgeLzXn7k8/YHA4A4CTcUhqj/D0oi1gI4pqihoMiSqUyPj7eTSIdN2seh8f7euPG9cu/uKM23tOYeCwUQSBdbczSmWOCPCU8zvUSQ57BorY6+gc/4dcxbtw4d3f33bt3i0SiadOmVYhFzMDQQIqLi6lMBkuWLGnfvv1rr722adOmiIiIKVOmPGvT/r0wPlhVsmTJks8++4x6zeFwkpOT6xogtDI///zz9OnTKx8f/Pr4D5d8y0KQXgFSJ0FS/lgUoRK37r6PBQ5VqVSff/759evX44aN+L9ZM8Tu5bGkVWVmwmry8fEBgNzc3LCwMNdWWVlZ4VQAaAAAIOGxbGcViq7kGiwywaOAilk6s787v66poKuC8cFqnuzcufO7774zmUzjxo37/PPPWax6/rtLqeR6JNneyz3PYDU5nIHuAlrQAMC5c+f69+9P1+fxeBaLhcqK+OO69XhQm5DW7a6fP711xedLftvj4eUTKnHr5iuhm9+6deuDDz64efNmx44dN27cKJPJRo4cmZqayuFwVu/8y69LDyrpDQJwuVBbanGcT/hDHhDUvleMO5fd3UdypVBtJ5EQiaC7rwcCgJOkBcPpUx0jyLMXkob0jwWArjH9563dwuHx5Hy2xo5zWGjfQE8PPocSW1wWKhdwFSabgM2KDa5XCs60NMjLAwBITIQffyR5PGL37vKvvX9/EFaUkgwMFfD39y8uLp42bdratWsB4OrVqz179mSz2XTKBIZnAPlicevWLWoLdMeOHS9evNiQroqKir7//vs+ffoMGTLk+PHjjWLeX3/95frlu7u7AwCLxfrqux8OZhavWrdpzJgxG/YeSchQJGQU31UbjzwoSchQ/PDngYEDB86ZM0ev15MkST0EA4BQKNy1axfdudVqLSkpoV5jGOYqsEJDQzEMo2uaHc6TOSqN1UEVy+zYyRyVwYaRTx2VSmU2m5/+uAzVcO7cOVfHoxUrVtSmVWlpaWJiYnZ2tuvBUzmlezMUN/MUJElaMOfxbOW+ewr6xCNJ0uFwuGYw/PDDDx+ZkXTph8NnFm7cyeHxAGDGx3OPZpWmFOuIhxUqnOT+/v7Dhz8KbYCyWH8n33QS5dVxgsjUmGxO/ESOMjFbacVwgiAe5BUczSo9k6vCcAInyKQCzaH7JXqbgyRJB06cyVMfziwOb1tuXpfesbvv5CdkKPamF+pcPsIdVVlChiIhQ/H3g1KTw1mvr9yFTZtIAMLNzfUHy/Bi8/3331e+NSMIUqdOqN9s9UcYnjIv1LeP47jrxLtcLm/IzVuhUDgcjprr1QUMw+jYpL6+vqmpqVlZWRqNhiCI/gMHAYCbu/jnxEt/3MhSme0kSeqt9u1X0jadv+Eh9waAuLg4s9ns7v4o/ZlIJDKZTCRJHj9+fMKECRMnTkxJSaHGSktL69u3r0QiiYmJSUtLczXjvtaUkKE4mFmssToMNuzw/ZKEDMVdtbFxP2xtYARWM2TRokWuF/oBAwbU2GTXrl08Ho+6KyxbtowkydLS0kGDBokkHv1HxC9cuJCqZsGcJSZbhbYqlWrevHljx4798ccfK6iKn9Zv8PL2RlF05MiROp3OijkJl3cLqLQwLlSYtV23bl1lU21O3IrhJEkSBJGfn292OB14ea9pqrKEDMWh+yUqs/1MnjohQ3EmT33q7DnqqvKfGXMpIbUnvShL9+ikvVGip44fzCzRWht8xWAE1nPO8ePHW7ZsGRERoVAo6tdD69atAYDNZtepVWVNRvmE3Llzp35mMDScF0pgVdyYA3Dr1q1699YUAovi+vXrp06dooQRRXp6Om2zzNevRYfOWVlZJEmmpaUFhLf0D3ukGm/evFnhM2ZnZx88eJCecuDz+dSnxgnXm1HFYrraSGmsQ/dLEjIUqaWGpvikNcIIrGZIhXnW2bNnV1/f6XRSwW8pUBQtKCgYP368ayfnzp2rtz1VSY0KM1gBAQHR0dGug/7888/VdEsJrAoH6ekoSl1R2uvPP/+k1NX2q+nD3/ngz1u5CRmKHL2ZIMlrxXpq7oqSWfQEWP25fp384gvDp58yAut5ZPLkya5nYHJycl17oPM+USdnx44doWpcFRVUElhCoRAANmzY0PDPxVA/XqhdhDKZrF27R5Gd/fz8WrVq9QztqYpu3boNHDhQ6OJXIZfLaYWkKSnOunPr1KlTAODn56cuKlDkZFFvyWSy9u3bu0YQ7dmzZ2ho6ObNm8mHTls2m23nzp12J3E6V52tt1AHTQ5nYo5KYXoU27CtTNRCKsQI0oETQWJBZ28mJTNDOfHx8fPmzePz+QiCjBw58ssvv6y+vt1up/MuAwBBECqV6tatW651UlNT621PVeGg2Gz2gQMHevbsyePxIiIiDh06VCFaSn5+fl3Hau0pov0LO8rdOSgCAP0Gxw175wNTmeGrSWMPbduYe+E4C0HS1aZiky3XYKH8rrr6SNrIRA6cSC2tGEm1bnTrBp9/bnLJhM3wHPHrr4/FZnvppZfq1Hzv3r3btm0DgE2bNlFZMm/fvl3N/buq4EEMzYQXSmChKLp///7+/fuLxeKePXsePnyYz+c/a6NqhVwur7C64eHhAQCenp4bNmygPoVYLN66dSuHwzl48OD06dP79es3bdq0w4cPIwhCObbT+Pj4GByYCcNTSw25BovJ4TxfoLVguNJsp+sYHc4iY3ng+BKTTcdEXmB4CIIgK1euNBgMJpNp//79rkvST8TNzU0mk7kecTgccXFxdJHD4QwYMKApTO3UqdOVK1dsNhsVlMTVXx4A+vXrV6fenAR5sVBrdeJ8NgsArih0VAgGH09pFwn37pG/2oeH/Pzzz3P/+350gLRvkKe/iN/dV0J7tXeQu3f2Fvfwlzbax2N4zrHb7TVXeoharY6PjweA+Pj4JwehrSOUe3vfvn0b3hVDPXlaU2XPH3VaIlSpVEqlsiHDnT59mt5M169fP7vdTr+l1+uvXbvmuqRYgYKCAjrtYO/evY1GI0mSJSbb/nvFezMU1CLgtWI9Xb/MjlHu86mlhnua8rXCRnAfqTvMEuGLQYVczn/88YfVap09e3Z4eHhERMT27dubYlCr1Xr9+vXS0lL6CIZhc+bM8fPzCw8PX7t2bfXNKywREgR59qHflQMn0pTl/lhG+6Olujt37kRHR7PZ7M6dOzdwD031FBUVMUuEzyMVVvSWL19e+7bUptGgoKBqOqyA65og4+TeDGG+/SqppcCy2Wzx8fHUqTxy5EiLxVLvEQsLC7dv3/7333/X49pqt9uPHz9+5MgRHMfpg9k6M+VNci5P7Vo5R2929bu6qzYmZCjua6sUcE0HI7BeDGbOnElf9D09PV0fNux2e3FxcaOPeOHCBblcDgAsFotyq68rBEGcOXPm9ddfj4yMnDlzpsFguK81nX3od0WS5B1VWVKBht6HSBBEmzZt6I8pl8ureeZ5ImlpaePHj+/Xr9+yZcuqv7YwAuv5ZdCgQWw2m8vlLl68uPatqGxsXC633uNSDznTpk2jisnJyVB3T3mGxoURWFVSS4H1zTffuD5SLF269CnY9kRcwzSQJGm0Y38/KKUE1t4MRY7+MR2jttgrFJ1O57x582QymUAgGDRoUGX/36aAEVgvBna7/auvvurTp8/YsWMr7CxpIoHlmgwKQZCMjIy69mAymVwn3kaNGkWS5ONbQR4rVsjkAwA3btyo/XBqtZpShBSzZs2qpjIjsP5VUI7tCII0cNs7dWotWbIkISGBeub/8ccfG9FOhrrCCKwqqaXAeu+991yvuRMnTnwKtj0RV4FldjiPZpVSK4PUWmFChiJbV92v9+OPP3b9IJGRkU/BZkZgvfA8UWDt2rVr9OjR77zzzrVr10iSdDqd2dnZ1Z8JOEE6XfSORCplsR45v1MewVFRUdU7Bbty5swZ1xMeRVEqm2FVEAThGrLL29v7idPVToJILTU4Hk4kO3AitdTgxAnXTM8A4OvrW81YjMD6V0HFZahAXeNgkSS5detW1wh27733XlNYy1B7Xign92fCiBEjXIsjR46sXMdJkIRLxHyMaPKEyVYn7sAJKgq8j5DXM0CKIojViVfT5LfffnMtpqSkWCyWJjaT4d/Ili1b/vOf/+zbt2/btm19+/bdvXt3aGhoeHi4t7f3li1bntiEBEgp1l8p0uEkCQA4QX6z69Ds79dRGosKDAEA//zzT3x8PFm77BTh4eGu+xNbtmxZfcB6BEH27dsXGxsrFAq7d+9++PBhgUBQuVqGxvRAZ04q0GI4gRFkUoHmgc58V2N0zaMAlUJ2MfybuXfvXuV7cz12CL7zzjuEy0NIhS2NDE8fRmA1lBEjRvz222+9e/fu1avX5s2bX3vttQoVcIK8XKS9/PDegOFEUoE2KadEqVQ1igFHjhzp1q2br6/vBx98UFZWvkVcJuAODJHTOXZ8hbxBofL28uq2gwUGBroWw8LCGpLBpqysTKfT1bs5wwvM9u3b6ddU/sHCwkIAMJvNU6ZMoRJoVsCJE0aHk0q848CJS0VaWUh41x69Xho6dOTIka63oszMTNeYEdUQFBS0bNky6iQPDAys8IDxRFq3bn3u3DmTyXTt2rWoqKgn1mnrKfJy4+ps2IVCbVKBRmfD5AJuO5l7VFTUokWLaEfm9evX18ZIBgaG5xcmF2GVFBcXy+VyDodTc9VqceDEhQKtwY75CHndfSVJ+Rojht+6dGHltHdefeWV//3vfw2JJfHgwYMOHTo4HA6q+Prrr+/evbt+XV2+fPnll1+mJJqXl9fZs2ddF0Rqj9PpnDx58o4dOwiCGDVq1M6dO6sRakwuwhceh8Oh1WpdvZ3Gjx//xx9/0EVqKYQunj9//okbyx04kVSg1dsxNoo4CdKdx+4XJGOhyJUCzaaf1vz9+zaVotDNXbz4l+1TxwwT82pOY0+SZGFhoaenZ2lpaUhISL3zLVYGJ8nLhTqlxQ4AMgG3T6AnGy1fuFEqlSUlJe3atav+wqJQKLy9vasKAMbAwPBcwMxgNTlcFto3yFPC45Sa7YnZKkpdrZj6tt1q3bt375o1axrS+aVLl2h1BQBJSUn17io6OrqoqOjcuXPp6elKpbJ+6goANm7cuG3bNmoz4759+1atWlVvkxheSJYtW0avlw0fPvzVV1+l3woMDIyMjHxiKy4LjQ6UshDESZAA4MBxm5NQGG2lVmzE+1NX7fk7MDTs+4Sjrbv3TNcYa2+MUCgMDw9vRHUFAAQJTrJ8Uq2Ce4C3t3fnzp0b/tjGwMDQ/GGekJ4GXBba09/jRK4KJ0mn3fbdRx84bOVB1TMyMhrSc7du3VAUpZdIunbt2pDeRCJRbGxsQ3oAgLt377oW79y508AOGV4wwsLCMjIyUlJS3N3dO3bsaDAYFi1adPny5fDw8GXLlj3RsQkAcJK8VmLASRIBIAHsTvJ8gaalVEgVpXKvbWeuGuyY3I0b4evxlD+RKxhBXizUaq2YXMBFEFBZHBcLtTFBMg6K1NyYgYHhBYKZwXoaYDhxtVhPkoAiCJvHn/Xtzxwul3rriU7xtadTp06bNm2iMu0MHDjw66+/bgx7G8TQoUOrKTIwAACHw4mOjqaCKEokkp9++iklJWX37t1V5bYqX3Qz2yU8TlyYl5TPAQAHTtxVGxEEAt35OEEa7JiUz+kd8Gg97plwX2vSWh1yAbd3oGfvAE/KHytTa3qGJjEwMDwTmpHAys/PX7BgQadOnYRCYcuWLWfOnFk58MzzCEYQSYVavQ2TC7hDwrwkPE732IEr/rcvpm/fbdu2jRo1qoH9v/vuuyqVymazHTlyxNvbu1FsbghDhw7dtm1bVFRU165d16xZM2nSpGdtEcNzD0kCCaSEx+kb5CnismMCPansNADARlGj3Um9NjqcZQ9fPyvaytzbyd17B3qyUYSFIr0DPNvJ3dvJRDW3ZGBgeLFoRk7u/fv3VyqVq1evjomJuXv37qRJkzAMu3HjxrPygG4sJ3cnQV4u0pIkUNdcO04kFWhFXFYPPynSqE/aNpvNYDBUyEvYzGGc3F94Kju51w/KmYnLQgGgoMyaUqxHALyEvFKzHQCkAo6PGy9DY2KjSKBY0N2nfP+s0mIvsztbSh8lVtfr9bNmzTp27Jivr+9HH33UPB8AGCd3BoYXgGY0gxUXF/fPP/+89NJLQqEwKipqw4YNmZmZe/bsedZ2NRQ2ikQHePYJKl+54LHQ2CDPRldXDAzPO67bNSrDRhFKXZXZnSnFegSBHgHSUEm5NHfgZDuZe3uZu5Mgc/WWNJURAJQW++VC3S1lmd4ll/mMGTN+++03pVJ569atKVOmPHjwoCk/EwMDw7+XZiSwFi5cKBQ+etCk9hnl5eU9O4saDTaKsFz0FIeFMuqKgYHm8OHDQUFBPB4vKiqqxm0fYh67vdy9h7/UX8QPcOdHB0hbSkURPhIEgbZyURsPPu6wZ2pNv52+crlQh5NkF2+xB//RPPTZs2fp1xiGXbp0qYk+FAMDw7+cZiSwKnDs2DF4KLNeVGw2W1JSUnp6+rM2hIHhMQwGw9mzZ3Nycp7CWEql8o033qBijaakpIwfP77GJm1kIn8R3+FwHDp06OS+v7wJs9ytfNfIRxPf+L/Xh5bptKKAEEpdtXBZHwQA1wChKIp27969UT8NAwMDQznNdI1frVYvXrw4KChozJgx1JFbt26lpqa61unSpUuLFi2azgaLxWI2m5suYs3du3dHjRpFZeh87bXXNm/e3MBgPDabjbK5kQx8GtDZ3J7aiK6zpAxP5MSJE6+99hoVcnbevHkrV65s0uHS09OtVitdvHnzptPprNH9yGKxDBw46Nr1a04ME4vFx48fj+rZMyc7++TJk+0ie3J5PKqa2mCsILDWrVtnNptPnDghl8s//fRTaicjA8PzjsViGThw4D///ENF7VmyZMmiRYuetVH/dpqjwMIwbOzYsSqVKjExkQ6Kk5WVlZiY6FpNJBIFBAQ0nRl2u91ms+F4dfn7GsKCBQvo/Od79uwZMWJEvSMa2Gw2nU7n6elJ2dx4NjY5NpsNRVEUfXozqYzAqpGpU6fSOZe+/vrriRMnNqkK6dq1q1QqpRMrxcbG1sa5+49du6Jff3Pwu9NWzXivrKxs6bJlC9ZvB7aoS59+837ewuXxD2xZPyh+XJHE447a2MElSZSvr+/x48edTieLxaKmzRgYXgDi4+OTk5N9fHzKyspcn1gYniHPRmBduXIlOjqaLrpOYJAk+fbbb585c2bnzp39+/enj48ePXr06NFP00iHwyGTyZpuBkuleiwXodFolMlk9ehn5cqVX375pc1m69y586+//tq2bdtGMvBpQJIks4uwuVFcXOxaVCgUTSqwJBLJ33//PWfOnPv37/fp02fdunW1aWVzYGHtOga2aDXnh40/zps5+L0ZJWa7hMtZ8ONGFo+/dflnx37fFtu5HbfnwEyNKdCdL+E99kNms9nNZwM1A0PDOXr0KPVCIpEwAqu5UDmJ97Plo48+AoC1a9c+a0NIhULhcDiarv/vv/+e/i9IpdLCwkK73f7JJ5+0adOmR48eBw4cqE0nycnJrv/NgQMHNp3BTYFKpTKbzc/aCobHmD59On1GtWjRwmq1NqQ3u91eXFzcWLbRKJXKth06rTl8NiFDsf1qekKG4mSOyu7ETQ7n1fu5hw4dys3NJUlSb3MUlD3ZfoIg8vPzG92wRqGoqAjDsGdtBUPduFWqT8hQFJWVuz1cUWj33ivWmGu+ibjeC2ioHJ31QCwWA8CSJUvq15yhEWleS4QrVqz44Ycfli5dOm3atGdtS5Mze/ZsmUy2b98+Ly+v2bNnBwQEzJ8/n87cFx8fn5KS0qVLl+o7qZCXJjU1dfXq1bGxsREREU1lN8OLzurVq1u2bHnu3Lng4OB58+Y1JBl50+Hl5XXmZOJvf/xJhLUQiSUsBOkTKOWyUC4LolqGQMsQqpqEx6kwd+WKyon4ESQd+T1HbwmRCJgtvgz144HeAgBXFIZe/lBotBUZbQBwpVg7tEUNsQlnz549e/ZsutimTZvMzMzGzY/J8ExoRoFGt27d+u67786dO/ebb7551rYANF6g0drTvXv3Gzdu0MXvvvvu448/rr5Jfn5++/btKzi2IwiyYcOGDz74oEmsbFSYQKMvPI0VaLQyOEleKdKVmu1UOkI/Eb+nv7T6NDmJiYnnzp0LCQl58803i23ETWWZ3I1LZde5XmLINVjCPNy6PQxS+gxhAo0+jxgw7FS2GgCoExIAWCiMbOVXp04mTZq0bds2AMjPzw8KCurUqVNaWlpVlREEoRPR0kgkkrKyMsbJvTnQjMI0fPfddwDw7bffIi64rla88HTq1Mm12Llz5xqbBAcHHzt2LC4uLiwsjD5IkuRnn33W+PYxMDQNeXl5W7Zs2b9/f/WxRl0hSDJZoSs12yU8zqAwLzGPXWyyXVXoiKofGFetWvXSSy8tX758ypQpcXFxfkKuCCXVFselQi2lrvhsVispsweCoZ5IOJyXw+W0ukKgzupq7969lLratGlTUFAQANy+fbuaFajK6oqhWdGMnpCq0en/Er799tuioqJTp04JBIL58+cPHjy4Nq1iYmISExMvXLgQGxtLH6zN5sf79+8rlcqIiIjmuQbE8C/h1KlTw4cPp9xye/XqdebMmdqckDgJdidBZSfkstCYQNmFAo0DJwiSrGqNzzUV+sWLF5MvXWzdomUuyVdbHWqrg8dGY4I8RdxmdElkeO64qzbRCp8ERGG0+rsLatlWrVbHx8cDQHx8/Pvvv980BjI8VZrRDBaDl5fXyZMny8rKdDrd4sWL69S2T58+/fr1o4sLFy6spjJBEG+99Vbr1q1jYmJatmzJSFuGxiUtLe27777bvn17baKyff755/SmpytXruzbtw8ADAbD/Pnzhw8fvnjxYqPRWLkVB0VigmSUugIAPhuNDZJR6T6rGoj3MDgWXWQBKeaW+wCIOGw3NuP1wlB/kot1+WU2AEDKJT55RaFXGGu7oY9aRg8KCnJNENepUyekap5mgBuGesA8rjU73N3da65UCRRFDx48uH37dq1W269fP9cIF5X5+++/d+zYQb0uKiqaM2fO8ePH6zEoA0NlDhw48NprrzmdTgBYvXq1a2qaJ1IhchtVHDZsWFJSEgAcPnw4OTm5Qgw8Cg6KADySUzx2DTeb5cuXv//++9Tk7ujRo3v27Hkpp1hpsfDZLD4b1Vgdl4u0vQM8WdW7cTEwVIHCaINyvytf2h8rVVlWm0ksb29vHMe5XG5+fr7r8du3bzeRtQxPAUZgvThwudw33njDx6eGHSsAUFRUVE2RgaEhLF26lFJXAJCamnrw4MEhQ4ZUU3/u3LkTJkygvElatGgxZsyY7OxsSl1RnDhxQqFQ+Pv7N9Cwd955JzIy8sKFC8HBwa+88kq23qJ0onw2KzbIk8dGkwq0KosjVVnW3ffZO7kzPI8Ma+13Klv1crgXAEg4nEHh8uRC3ZAw7xobTpo0SaVSIQhCh9tleDFgBNa/keHDh8+fP5/+Mb/99tvP1h6GF5ga9yn/5z//CQkJOXLkiKen5zvvvCORSACAz+fTM1sikcjDw6NRjOnYsSMdNDXUw61Ia4gI8qL8rmKCPG+WlrWXixplIIZ/IRwASl1RSDic2qgrAKAyjpMk6Zpn4ok7BKtBrVZ7eT0affHixYsXL65rJwyNC7OC+2/E39//4sWL77///ogRIzZv3jx37txnbRFD/cnPz1+wYEGnTp2EQmHLli1nzpyp0WieoT0LFy6k4wt06dJl5MiRNTaJjo5eunTpxx9/7OnpCQASieSnn36iIqTweLz169c3RSAPFgSwzagAABBnSURBVIKE8wjaq52DolF+HnzGDYvhqXPv3r2G7xCUy+XMNsPmRjOKg9XcePpxsBqIzWYzGAy1WSJsPjBxsBpO//79lUrl6tWrY2Ji7t69O2nSJAzDbty48Qy/1du3bx87dkwmk40dO5bD4dQvDpZSqbx3717btm1dn8sbEZIkCwsLqc3wzQ0mDhYDwwsA8wNmYHi+iYuL++ijj6jFhaioqA0bNvTt23fPnj1vvfXWszKpU6dOdFC32oe2qoC3t7e3d61WWBgYGBiaIYzAYmB4vqkQkiM8PBwA8vLynpE5DAwMDAwAjMBiYHjBOHbsGDyUWS8YTqczKSkJw7DY2NgKQa0YGBgYmhuMwGJgeHFQq9WLFy8OCgoaM2bMUx46KytLLBY3kb8UAOh0ugEDBqSmpgJAy5Ytz549GxAQ0ERjMTAwMDQcRmAxMLwgYBg2duxYlUqVmJgoEJTHNty2bdvWrVtdq02aNOmVV15pxHFLS0vffPPNW7duoSg6YcKEr7/+2jXANIZhOp0OqSJ9TW2w2WwnT55MSEig1BUAPHjw4PPPP1+2bFk9etNoNIWFha1atRIIBGq1msvl1tuwpkOlUpEk2Xyc3J+vrTMMDM2E5vIDZmBgqJErV65ER0fTRdctwCRJvv3222fOnNm5c6drHP8BAwa4JgIHgNDQ0MYKK0UxZ86cW7duAQBBEDt27HjllVdc588cDgdBEPUe0Wg0Dh8+vHI8a5PJVI8+165dO3/+fAzDJBLJb7/91qFDh8b9KhoLq9Xq4eHRfAQWAwNDPWB+wAwMzw29evWqKq7Kxx9//Mcff6xdu3b8+PGux0NCQkJCQprUquzsbNdibm6uq4MUgiBcLreyy5TVaqWn2aphy5YtldUVi8WaMmVKXd2wCgsL582bR6XKMRgM//3vf//++++ZM2empKS0bdt2+fLlTf1F1R7qG2MEFgPDcw0TaJSB4blnxYoVP/zww9KlS6dNm/b0R4+Pj6dfc7ncYcOGVV//7t273bp1c3NzCwgIOHToUPWVK2QqDAwMnDx58rlz5wYMGFBXO4uKiih1RVFaWvrBBx9s3Ljx+vXrv//++5AhQ+gMPwwMDAwNhwk0WiVMoNGnABNotOFs3br13XffnTt37jfffPNMDCAI4scff9yzZ4+7u/v//d//DRw40PVdh8NRIdBo586d6UkpgUCQl5dXjWu8Uqns3r07lS6TzWYfOHBg6NCh9bMTw7DOnTtnZGRQxREjRhw8eNC1ws2bN7t06VK/zhsXJtAoA8MLACOwqoQRWE8BRmA1nI4dO965c6fCwWnTpq1du/aZ2FOBCgILwzAej+d62Tl//nzfvn2r6aG0tHT79u0mk2nUqFHdu3dviDHFxcVLly69f/9+r1695s6d26pVK6VSSb0lEAiKioqkUmlD+m8sGIHFUFdWrVq1ePFiOq4vm83+888/n/5uYgZXGIFVJYzAegowAuuFp/IMVr9+/c6fP0+9lslk2dnZYrG4UcYyOZx0bkESwILhQk6VuQVJkty9e/esWbNKS0vFYvGmTZveeOONRjGj4TACi6GuUBt1u3fvvnHjxqVLl+7fvx+a06TsvxPGB4uBgeGpsmvXrjFjxvj6+sbExBw7dqyx1NVtVdnpPLXK4gAAEuBasf5Mnlpvx6pp0rt378LCwqysrNLS0uajrhieb6ZPBw4HOBzIzHyawwoEgpSUlGvXrkVEROzbt2/QoEEA0LgBWRjqCiOwGBgYnip+fn4JCQnFxcUXLlyIjIxsrG5FHLaTIC8VaUvN9usl+vwyKwtBOGgNlzg2mx0eHs7n8xvLDIZ/O0lJ4HSC0wkFBbVssXr1aqQSaE2nbgUsFktERARdpBYH9Xp9nTphaFwYgcXAwPAiEObh1sVbjBPkxUJtnsHqxmHFBsuqWSJkYGgmzJ49m3ShdevWAMBiNejU/eqrrwBAJpM1jokM9YIRWAwMDI0MTpAOnHhUBHAST8PXM1wq9BSUR2ZvJ3Nn1BXD0+P99yEwEAID4e7d8iPjxpUf+fXX2nczadKkzMxMeBherlOnTpUnt2qc5frrr79KS0sB4Pjx4w38WAwNgRFYDAwMjQlOkpeLdBcKtJTGwgjyjom8WKhtao1FAtwoMWitDi4LBYBUpUFtcTTpiAwMj/j9dygqgqIiwB66/alU5UdqLbD27t27bdv/t3dvMU1teRzH12q5DC0Hi4DnMNrMiI6ihME4M9Ij3h11EjUm+mDQqCGa+GBi1BgNXhIe0PigiYkhPph4fSAx3qPBCOIDjAcTo0bUCcaE6oFYKFbwUgcq7HnoHFNp5VC6dvcGv58nuru6+fvPXvXH7u5ep4UQJ06ccDqdQojGxkbt2/r6+sJ38vr16+AFhbNmzZo6daqKfxuGiIBlRl1dXaWlpcuWLdu7d++7d++MLgeIghTCYhFd3YG6X30fej7f87z70Pv7r4rdQ0+Xu8tvS7TO/1PmX8ekBa/H8n0a6CJ3IB4G923Qjo6O4D17V61atWnTpqH9qkAgEFwE3el01tXVDW0nUIWvAZvR8uXLg3Pjxo0bDQ0Nt2/fNroiYLAsUhb+Mb2h9W3bx+5qt1fTRFqCKBo3OsEy9PWeB2OMPcnr757lzLAnWiem24UQzZ1+G58SIj78/v//MG2aCK5KXlMjFi4c/A6CtzJxOp0XLlz4sjE/P//JkyffeomUst9JLJvNpmlaenr6q1evoige+uAMluk0NzeH/uVRW1vb0tJiYD1AtKxS/v0nR4JFapqwSPmXFKl3uhJCjPsh5Z/js75cdzUx3b7gz5l/SOAtDsPAmDFjent7k5KS+gWjqD4itNvtnz9/TklJ8fl88S0fkfHuYzoZGRmhi+Da7XaT3F0aGKRAn/ZLq+9zn5ZgkX2a9p+PX13zrh+rlAM8BMyppKTE6/VKKd++fTvknWRnZ/v9/sTERP+Xc2kwGgHLdNLS0ioqKpKSkoQQycnJx48ft9vtRhcFDNbnPu3fv77x/TeQaUv6V86YrJREf5+ob/EFIl2QC4w0FRWiuFgUF4uffx7kK+7evSuE0DTNbrcP+T5YHo9HCBEIBEK/ZshiAMai+2ZUUlKybNmypqam3NzczMxMo8sBomC1yNSkBCnlzLGjEyzyHz/+8EurL9lq4XwSvgtFRaKoKKpXNDU1xf5rWfXOhAhYJpWVlZWVlWV0FUDUpBB/y3b09WlWixRCWKScbJM//Tha/6uwAMBE+IgQgGJSCGtInrIIQboC8L0hYAEAAChGwAIAAFCMgAUAAKAYAeubLl++3NnZaXQVUfB4PFVVVUZXEZ07d+643W6jq4COvF7v9evXja4igt7e3rNnzxpdRWQXL17s6uoyugoAMSFgfVNFRUV7e7vRVUTB7XYHFwodRs6fP//06VOjq4COWlpaTpw4YXQVEQQCgfLycqOriOzYsWNv3rwxugoAMSFgAQAAKEbAAgAAUIyABQAAoBgBCwAAQDFrWVmZ0TWYlJRyxowZw2uhZYfDMW3aNKOriIKUMi8vj0WBRjApZVpa2vTp040upD8ppdVqnTNnjtGFRFZYWGiz2YyuAsDQSVaIBAAAUIuPCAEAABQjYAEAAChGwAIAAFCMgAUAAKAYAUsIIV69erVnz578/Hy73T5x4sStW7cOsE5FTU2N/FpmZmbcSm1vb1+3bt3o0aNTU1OXLFky8DozUQ3WyTDqLXQS1TEQH2aYGuFM2CgAQ0bAEkKI9evXX7ly5fDhw+3t7ZWVlbW1tTNnzvT7/QO8pLGxUftNR0dHfOoMBAKLFy9+8eLFgwcPXr58mZGRMW/evNevX8c+WD/DpbfQzxCOAV2ZZGqEM1ujAMREg6aVl5d/+PDhy8O6ujohxJkzZyIOrq6uFl+HgLgJruX87Nmz4MP379+npaVt37499sH6GS69hX6iOgbiwCRTI5zZGgUgFpzBEkKIvXv3ht5QNCcnRwjx8uVL4yqK7Nq1axMmTJgyZUrwYWpq6oIFC65evRr7YP0Ml95CP2Y7BkwyNcKZrVEAYkHAiuDmzZvit3e3b5k/f35iYmJ2dnZJSUlra2t8Cnv69OmkSZNCt0yePLm5ufnTp08xDo4b0/YWcTOYY0BX5pwa4QxvFIBYELD66+jo2L9/v9PpXLlyZcQBycnJ+/btq6+v9/l8Z8+era+vd7lcXq83DrX5fL5Ro0aFbnE4HJqmdXZ2xjg4PszcW8TH7x4DcWDCqRHODI0CEIsEowswl0AgsHr1aq/Xe+vWrZSUlIhjZs+ePXv27ODPixYtunTpUkFBwdGjRw8cOBDHSocfeovBHAMQNAoYEb67M1gNDQ2hdwEIfUrTtA0bNty5c+f06dPz5s0b5A7z8/OdTue9e/fU1xomPT29q6srdEtnZ6eU0uFwxDhYb+bvLZRQPr/0YKqpEc48jQIQi+8uYLlcrtCL/EOf2rFjR2Vl5bFjx9asWRPVPqWM05rZeXl5z58/D93S1NQ0fvz4iH/jRjVYb+bvLZTQY34pZ6qpEc48jQIQk7h8V3EYOHjwoBCivLw82hc2NjZKKUtLS/Woqp9Tp06JsK+Xb9u2LfbBuhoWvYWuhnwM6ME8UyOcqRoFIBYELE3TtJMnTwohdu7cGfHZqqoqIcS5c+eCDzdu3Hju3Dm32/3+/fuamppJkyaNHTu2ra0tDnX29PTk5+e7XK7m5uaOjo7i4uKMjIyWlpaIdQ48OG6GS2+hn4GPgfgzydQIZ7ZGAYgFAUvTNC0vLy/83N6WLVuCz/YLAS9evNi8eXNOTk5iYuK4ceM2bdrU2toat1I9Hs/atWsdDofNZlu0aFHoPTn71Tnw4LgZRr2FTgY+BgxhhqkRzoSNAjBkXOACAACg2Hd3kTsAAIDeCFgAAACKEbAAAAAUI2ABAAAoRsACAABQjIAFAACgGAELAABAMQIWAACAYgQsAAAAxQhYAAAAihGwAAAAFCNgjRwfP37Mzc2dMWNGIBAIbrl165bFYqmoqDC2MMC0mDUAdMJizyPKw4cPXS7X9u3bDx061NbWVlBQUFhYePXqVaPrAsyLWQNAD9aysjKja4Ay2dnZqampZWVlRUVFu3bt8nq9VVVVNpvN6LoA82LWANADZ7BGGk3Tli5dWltb29PTU11dvXDhQqMrAsyOWQNAOa7BGmmklOvWrevu7i4oKOD/CWAwmDUAlOMjwpHG4/GsWLEiNzf30aNHDofD5XIZXRFgdswaAMpxBmtE0TRtw4YNycnJNTU127Zt27179+PHj40uCjA1Zg0APXAN1ohy5MiRXbt21dbWzp07t6enx+VydXd3379/PyUlxejSAJNi1gDQA2ewRo4HDx7s2bOntLR07ty5QoikpKTKykq3271jxw6jSwNMilkDQCecwQIAAFCMM1gAAACKEbAAAAAUI2ABAAAoRsACAABQjIAFAACgGAELAABAMQIWAACAYgQsAAAAxQhYAAAAihGwAAAAFCNgAQAAKEbAAgAAUIyABQAAoBgBCwAAQDECFgAAgGIELAAAAMUIWAAAAIoRsAAAABQjYAEAAChGwAIAAFCMgAUAAKAYAQsAAEAxAhYAAIBiBCwAAADFCFgAAACK/Q+8bL0POCWGogAAAABJRU5ErkJggg==" />
<h2>Model selection</h2>
<h3>Testing the sufficient condition for <span class="math">$X\to Y$</span></h3>
<p>For the sufficient condition we have to test whether <span class="math">$Z$</span> and <span class="math">$Y$</span> are dependent and whether this dependence disappears after conditioning on <span class="math">$X$</span>. If we assume linear models, these tests can be performed by standard least squares regression and testing for non-zero coefficients.</p>
<h4>Testing data generated by Model 1a</h4>
<p>First we test whether <span class="math">$Z$</span> and <span class="math">$Y$</span> are dependent by regressing <span class="math">$Y$</span> on <span class="math">$Z$</span>. Note that by construction <span class="math">$Y$</span> and <span class="math">$Z$</span> are samples from random variables with zero mean and hence we don't include an intercept in the regression:</p>
<pre class='hljl'>
<span class='hljl-n'>yg</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>lm</span><span class='hljl-p'>(</span><span class='hljl-nd'>@formula</span><span class='hljl-p'>(</span><span class='hljl-n'>Y1a</span><span class='hljl-t'> </span><span class='hljl-oB'>~</span><span class='hljl-t'> </span><span class='hljl-ni'>0</span><span class='hljl-t'> </span><span class='hljl-oB'>+</span><span class='hljl-t'> </span><span class='hljl-n'>Z</span><span class='hljl-p'>),</span><span class='hljl-n'>data</span><span class='hljl-p'>)</span>
</pre>
<pre class="output">
StatsModels.TableRegressionModel{LinearModel{GLM.LmResp{Vector{Float64}}, G
LM.DensePredChol{Float64, CholeskyPivoted{Float64, Matrix{Float64}}}}, Matr
ix{Float64}}
Y1a ~ 0 + Z
Coefficients:
─────────────────────────────────────────────────────────────
Coef. Std. Error t Pr(>|t|) Lower 95% Upper 95%
─────────────────────────────────────────────────────────────
Z 0.676864 0.146003 4.64 <1e-05 0.388952 0.964777
─────────────────────────────────────────────────────────────
</pre>
<p>From the small <span class="math">$p$</span>-value we can reject the null hypothesis of no dependence between <span class="math">$Y$</span> and <span class="math">$Z$</span>, and hence we conclude that the first test is passed.</p>
<p>Next we test whether <span class="math">$Z$</span> and <span class="math">$Y$</span> become independent after conditioning on <span class="math">$X$</span>, that is, after regressing out <span class="math">$X$</span> from <span class="math">$Y$</span>:</p>
<pre class='hljl'>
<span class='hljl-n'>yx</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>lm</span><span class='hljl-p'>(</span><span class='hljl-nd'>@formula</span><span class='hljl-p'>(</span><span class='hljl-n'>Y1a</span><span class='hljl-t'> </span><span class='hljl-oB'>~</span><span class='hljl-t'> </span><span class='hljl-ni'>0</span><span class='hljl-t'> </span><span class='hljl-oB'>+</span><span class='hljl-t'> </span><span class='hljl-n'>X1a</span><span class='hljl-p'>),</span><span class='hljl-n'>data</span><span class='hljl-p'>)</span>
</pre>
<pre class="output">
StatsModels.TableRegressionModel{LinearModel{GLM.LmResp{Vector{Float64}}, G
LM.DensePredChol{Float64, CholeskyPivoted{Float64, Matrix{Float64}}}}, Matr
ix{Float64}}
Y1a ~ 0 + X1a
Coefficients:
────────────────────────────────────────────────────────────────
Coef. Std. Error t Pr(>|t|) Lower 95% Upper 95%
────────────────────────────────────────────────────────────────
X1a 0.826969 0.0589218 14.04 <1e-30 0.710778 0.94316
────────────────────────────────────────────────────────────────
</pre>
<p>The residuals of the regression of <span class="math">$Y$</span> on <span class="math">$X$</span> can now be tested for association with <span class="math">$Z$</span>:</p>
<pre class='hljl'>
<span class='hljl-n'>data</span><span class='hljl-oB'>.</span><span class='hljl-n'>residYX1a</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>residuals</span><span class='hljl-p'>(</span><span class='hljl-n'>yx</span><span class='hljl-p'>)</span><span class='hljl-t'>
</span><span class='hljl-n'>ygx</span><span class='hljl-t'> </span><span class='hljl-oB'>=</span><span class='hljl-t'> </span><span class='hljl-nf'>lm</span><span class='hljl-p'>(</span><span class='hljl-nd'>@formula</span><span class='hljl-p'>(</span><span class='hljl-n'>residYX1a</span><span class='hljl-t'> </span><span class='hljl-oB'>~</span><span class='hljl-t'> </span><span class='hljl-ni'>0</span><span class='hljl-t'> </span><span class='hljl-oB'>+</span><span class='hljl-t'> </span><span class='hljl-n'>Z</span><span class='hljl-p'>),</span><span class='hljl-n'>data</span><span class='hljl-p'>)</span>
</pre>
<pre class="output">
StatsModels.TableRegressionModel{LinearModel{GLM.LmResp{Vector{Float64}}, G
LM.DensePredChol{Float64, CholeskyPivoted{Float64, Matrix{Float64}}}}, Matr
ix{Float64}}
residYX1a ~ 0 + Z
Coefficients:
──────────────────────────────────────────────────────────────