hmb_mp_phenotype.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "871cd589",
   "metadata": {},
   "source": [
    "# HMB and MP Phenotype Derivation on UK Biobank RAP\n",
    "Notebook for deriving Heavy Menstrual Bleeding and Menstrual Pain phenotypes for use in a Genome Wide Association Study (GWAS) on the UK Biobank RAP platform.        \n",
    "*Cell outputs removed in compliance with Biobank data protection rules*\n",
    "\n",
    "Author: **Adam El Kholy**    \n",
    "Project P.I: **Dr. Gemma Sharp**    \n",
    "Department of Psychology in collaboration with the Exeter Genetics Group,           \n",
    "University of Exeter    \n",
    "Last updated: 27/01/2025"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "058cee82-0fe5-4f46-afe6-6cc558f93816",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on database app103356_20241205153134...\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from pyspark.sql import functions as F\n",
    "from ukbython import ukbython\n",
    "ukb = ukbython()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12fbece3",
   "metadata": {},
   "source": [
    "# Heavy Menstrual Bleeding (HMB) phenotype"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca2b5807-45d0-48a5-89b1-c0cb4bb9f218",
   "metadata": {
    "tags": []
   },
   "source": [
    "## ICD9, ICD10, Read2 and CTV3 codes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "725cd91e-b16d-4e2d-bf66-b15672d14f8e",
   "metadata": {},
   "source": [
    "Getting ICD9 eids and dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d2fc2024-0d6a-435b-a0b0-38fe6b0018a3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 2 ICD9 codes...\n",
      "Operation complete: Found 1592 matching records in 20.75 seconds.\n"
     ]
    }
   ],
   "source": [
    "icd9s = [\n",
    "    \"6262\",  # excessive or frequent menstruation\n",
    "    \"6270\"   # premenopausal menorrhagia\n",
    "]\n",
    "hmb_icd9 = ukb.get_icd9(icd9s)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5de18745-ecb7-4643-ba37-e5c63d9fcd49",
   "metadata": {},
   "source": [
    "Getting ICD10 eids and dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "243ff78d-6a4e-4630-984d-994e74557680",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 4 ICD10 codes...\n",
      "Found 14715 matching hospital records\n",
      "Found 0 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 14715 matching records in 32.05 seconds.\n"
     ]
    }
   ],
   "source": [
    "icd10s = [\n",
    "    \"N92.0\",  # excessive and frequent menstruation with regular cycle\n",
    "    \"N92.1\",  # excessive and frequent menstruation with irregular cycle\n",
    "    \"N92.2\",  # excessive menstruation at puberty\n",
    "    \"N92.4\"   # excessive bleeding in the premenopausal period\n",
    "]\n",
    "hmb_icd10_hesin = ukb.get_icd10(icd10s)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47a959ed-63a0-4f1e-bda6-3370994488d3",
   "metadata": {},
   "source": [
    "Get GP clinical eids (Read2/CTV3 codes) and dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9a608e88-e462-419c-bdb7-8bc0ace67c69",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "Read2 = [\n",
    "    \"K5A0.11\",  # climacteric menorrhagia\n",
    "    \"K592.00\",  # excessive or frequent menstruation\n",
    "    \"K592z00\",  # excessive or frequent menstruation nos\n",
    "    \"1573.11\",  # h/o: heavy period\n",
    "    \"573.00\",   # h/o: menorrhagia\n",
    "    \"K592012\",  # heavy menstrual bleeding\n",
    "    \"K592011\",  # heavy periods\n",
    "    \"K592000\",  # menorrhagia\n",
    "    \"K5A6.00\",  # perimenopausal menorrhagia\n",
    "    \"K5A0.00\",  # premenopausal menorrhagia\n",
    "    \"K593.11\",  # pubertal bleeding and menorrhagia\n",
    "    \"K592.12\"   # hypermenorrhoea\n",
    "]  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "353241a4-d821-4737-ac7f-1dacc2ccc475",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 12 Read2/CTV3 codes...\n",
      "Operation complete: Found 14957 matching records in 5.64 seconds.\n"
     ]
    }
   ],
   "source": [
    "hmb_gp_clinical = ukb.get_gp_clinical(Read2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "63fe3d5d-9f4d-4c87-bca5-aacf4008962c",
   "metadata": {},
   "source": [
    "Checking ICD10 death records: no entries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "301d5a08-2874-48a5-9aae-8e134f303c2a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# hmb_icd10_death = ukb.get_icd10_death(icd10s)\n",
    "# hmb_icd10_death.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6e640fb-3023-4e3d-9c64-01ee7d5b6f19",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Tranexamic acid phenotype"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ec06a4b-55e8-4756-9579-80492e212c24",
   "metadata": {},
   "source": [
    "Treatment/medication code: field code: 20003  \n",
    "Tranexamic acid n=699    \n",
    "code: 1140861832"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cbf2ecf-8842-4776-afe2-8ded673caaa1",
   "metadata": {},
   "source": [
    "p20003_i{i}_a{a}\n",
    "for  \n",
    "0 to 3 inclusive instances, \n",
    "0 to 47 inclusive arrays with \n",
    "no dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "edef70f9-17ab-4022-85d3-8a7791911325",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# all fields potentially containing a treatment code\n",
    "columns = [f\"p20003_i{i}_a{a}\" for i in range(4) for a in range(48)]\n",
    "\n",
    "# extract eids with treatment code for tranexamic acid\n",
    "query = f\"SELECT DISTINCT eid FROM `{ukb.database}`.`participant_0032` WHERE {' OR '.join([f'{col} = 1140861832' for col in columns])}\"\n",
    "\n",
    "tranex_df = ukb.spark.sql(query)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "445e40c9-8675-4dbc-8d82-8bf6075ec7e4",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Endometrial ablation phenotype"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe6cc1f1-adad-4f76-b73a-60677cb95039",
   "metadata": {},
   "source": [
    "Endometrial ablation field code = 20004   \n",
    "operations code = 1360"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9034c5f5-1e39-4421-bf31-6dd1b9896959",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# all fields potentially containing a self-reported operation \n",
    "columns = [f\"p20004_i{i}_a{a}\" for i in range(4) for a in range(32)]\n",
    "\n",
    "# concatenate all fields into one query to check for endometrial ablation\n",
    "case_statements = [f\"WHEN {col} = 1360 THEN '{col[8]},{col[11:]}'\" for col in columns]\n",
    "case_condition = \" \".join(case_statements)\n",
    "\n",
    "# extract (instance, array index)  for date querying\n",
    "query = f\"\"\"\n",
    "SELECT eid, \n",
    "       CASE {case_condition} \n",
    "       ELSE NULL \n",
    "       END AS idx\n",
    "FROM `{ukb.database}`.`participant_0032`\n",
    "WHERE { \" OR \".join([f\"{col} = 1360\" for col in columns]) }\n",
    "\"\"\"\n",
    "\n",
    "endo_df = ukb.spark.sql(query)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ef74f1c-2229-41aa-ab4d-26dae596921b",
   "metadata": {},
   "source": [
    "Extracting fields for date of first diagnosis calculation:\n",
    "- p20010: date of diagnosis\n",
    "- p20011: age of diagnosis  \n",
    "\n",
    "Extracting instance and array fields (4 instances with 32 arrays each):\n",
    "- p20010 fields i0_a0 to i0_a9 are in p_0033, the rest are in p_0034\n",
    "- p20011 fields are all in p_0034"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "750335eb-bf0d-4431-80f3-e70efe65e1dd",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# field names for querying p0033 and p0034, defined as above\n",
    "p_33_selections = [\"eid\"] + [f\"p20010_i0_a{a}\" for a in range(10)]\n",
    "p_34_selections = [\"eid\"] + [f\"p20010_i{i}_a{a}\" for i in range(4) for a in range(32) if int(i) != 0 or int(a) > 9] + [f\"p20011_i{i}_a{a}\" for i in range(4) for a in range(32)]\n",
    "\n",
    "# querying p0033, p0034, and date of birth register\n",
    "p_33_df = ukb.spark.sql(f\"SELECT eid, {', '.join(p_33_selections)} FROM `{ukb.database}`.`participant_0033`\")\n",
    "p_34_df = ukb.spark.sql(f\"SELECT eid, {', '.join(p_34_selections)} FROM `{ukb.database}`.`participant_0034`\")\n",
    "dob_df = ukb.spark.sql(f\"SELECT eid, p34 FROM  `{ukb.database}`.`participant_0001`\")\n",
    "\n",
    "# inner joins to create master dataframe\n",
    "main_df = endo_df.join(p_33_df, on=\"eid\", how=\"inner\")\n",
    "main_df = main_df.join(p_34_df, on=\"eid\", how=\"inner\")\n",
    "main_df = main_df.join(dob_df, on=\"eid\", how=\"inner\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "44e64452-be88-4ec8-9a40-0ab0d54f4383",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# convert spark DataFrame into Python list\n",
    "endo_list = main_df.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8847f1f9-a079-459b-aa1e-98cbff75afc9",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "endo_ages_list = []\n",
    "\n",
    "# enumerate endometrial ablation eids\n",
    "for entry in endo_list:\n",
    "    # extract instance, array indexes\n",
    "    i, a = entry[\"idx\"].split(\",\")\n",
    "\n",
    "    # get date of birth\n",
    "    dob = entry[\"p34\"]\n",
    "\n",
    "    # get age and date of diagnosis\n",
    "    age = entry[f\"p20011_i{i}_a{a}\"]\n",
    "    date = entry[f\"p20010_i{i}_a{a}\"]\n",
    "\n",
    "    # convert date to age\n",
    "    date_to_age = round(date - dob, 1)\n",
    "\n",
    "    # validation for calculating correct min value\n",
    "    if age <= 0:\n",
    "        age = 10000\n",
    "    if date_to_age <= 0:\n",
    "        date_to_age = 10000\n",
    "    first_diagnosis_age = min(age, date_to_age)\n",
    "\n",
    "    if first_diagnosis_age == 10000:\n",
    "        first_diagnosis_age = None\n",
    "\n",
    "    endo_ages_list.append([int(entry[\"eid\"]), first_diagnosis_age])\n",
    "\n",
    "# print(\"\\n\".join([str(e) for e in endo_ages_list[:10]]))\n",
    "# print(len(endo_ages_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3d6b9142-a9be-4e2e-a177-ba1c6675f7db",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "endo_ages_df = ukb.spark.createDataFrame(endo_ages_list, schema=['eid', 'first_diagnosis_age'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "117473bd-a923-4c90-9b27-05493126068e",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Final HMB phenotype"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "353d295b-7ad6-4e61-bcaf-e69e74a4dbb1",
   "metadata": {},
   "source": [
    "Derived variables:   \n",
    "• hmb_ever <- any diagnosis ever recorded vs no diagnosis ever recorded    \n",
    "• hmb_incident_ado <- any diagnosis before or equal to age 19 vs no diagnosis before or equal to age 19    \n",
    "• hmb_incident_mid <- any diagnosis between age 20-40 (inclusive) vs no diagnosis between age 20-40 (inclusive)    \n",
    "• hmb_incident_meno <- any diagnosis between age 40-60 (inclusive) vs no diagnosis between age 40-60 (inclusive)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d4b7f5f6-7843-45e7-9b71-b6aa3cfec44d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# join all dates entries into one table\n",
    "all_dates_df = hmb_icd9.join(hmb_icd10_hesin.withColumnRenamed(\"diagnosis_date\", \"icd10_diagnosis_date\"), on=\"eid\", how=\"outer\")\n",
    "all_dates_df = all_dates_df.join(hmb_gp_clinical, on=\"eid\", how=\"outer\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ea24a40d-1b94-4b86-b754-dd345d5ece11",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# get the earliest date across all fields\n",
    "first_date_df = all_dates_df.withColumn('first_diagnosis_date', F.least(F.col('diagnosis_date'), F.col('icd10_diagnosis_date'), F.col('event_dt')))\n",
    "first_date_df = first_date_df.select(\"eid\", F.year(\"first_diagnosis_date\").alias(\"first_diagnosis_date\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6d30555c-a4b4-422f-8827-bfe3afc9882b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# join dobs with first diagnosis dates\n",
    "dates_to_ages_df = first_date_df.join(dob_df, on=\"eid\", how=\"inner\")\n",
    "\n",
    "# subtract date of birth from diagnosis date to get age at diagnosis\n",
    "dates_to_ages_df = dates_to_ages_df.withColumn(\n",
    "    \"first_diagnosis_age_unfiltered\", \n",
    "    dates_to_ages_df[\"first_diagnosis_date\"] - dates_to_ages_df[\"p34\"]\n",
    ")\n",
    "\n",
    "# replace ages <= 0 with null in a column \n",
    "dates_to_ages_df = dates_to_ages_df.withColumn(\n",
    "    'first_diagnosis_age_',\n",
    "    F.when(F.col('first_diagnosis_age_unfiltered') <= 0, F.lit(None))  # if value <= 0, set it to null\n",
    "    .otherwise(F.col('first_diagnosis_age_unfiltered'))                # otherwise keep the original value\n",
    ")\n",
    "\n",
    "all_ages_df = dates_to_ages_df.select(\"eid\", \"first_diagnosis_age_\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "0473bf4f-0ceb-466c-9536-effd1e640abd",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# finally get mininum age of diagnosis from [min(ICD9, ICD10, Read2 codes)] , [endometrial ablation]\n",
    "combined_ages_df = all_ages_df.join(endo_ages_df, on=\"eid\", how=\"outer\")\n",
    "final_ages_df = combined_ages_df.withColumn('diagnosis_age', F.least(F.col('first_diagnosis_age_'), F.col('first_diagnosis_age')))\n",
    "final_ages_df = final_ages_df.select(\"eid\", \"diagnosis_age\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "733f1e6f-4c47-4136-8d1d-1b529c4920b2",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# finally join tranxamic acid participants (for whom no dates are available)\n",
    "hmb_phenotype_df = final_ages_df.join(tranex_df, on=\"eid\", how=\"outer\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ab5601d-80dd-4182-8a34-9e1db1603576",
   "metadata": {},
   "source": [
    "## Filtering female participants only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "cde0a5ef-6eba-4f14-83c5-07697d30c28b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "query = f\"SELECT DISTINCT eid FROM `{ukb.database}`.`participant_0001` WHERE p31 == 0\"\n",
    "females_df = ukb.spark.sql(query)\n",
    "hmb_phenotype_df = females_df.join(hmb_phenotype_df, on=\"eid\", how=\"inner\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a828b0d-029d-4ef9-b9e0-0eaec4799348",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Deriving HMB variables\n",
    "• hmb_ever <- any diagnosis ever recorded vs no diagnosis ever recorded    \n",
    "• hmb_incident_ado <- any diagnosis before or equal to age 19 vs no diagnosis before or equal to age 19    \n",
    "• hmb_incident_mid2 <- with a different age at first diagnosis range: 20 <= age <= 40          \n",
    "• hmb_incident_meno2 <- with a different age at first diagnosis range: age >= 40    \n",
    "• filter out over 60s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "91c97363-0d20-4aa6-a155-f88a7bd3729f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def derive_variables(df, name):\n",
    "    phenotype_df = df.withColumn(\n",
    "        \"diagnosis_age\",\n",
    "        F.when(F.col(\"diagnosis_age\") < 0, None).otherwise(F.col(\"diagnosis_age\")),\n",
    "    )\n",
    "    # age ≤ 19\n",
    "    phenotype_df = phenotype_df.withColumn(\n",
    "        f\"{name}_incident_ado\",\n",
    "        F.when(\n",
    "            (F.col(\"diagnosis_age\") > 0) & (F.col(\"diagnosis_age\") <= 19), F.lit(1)\n",
    "        ).otherwise(F.lit(0)),\n",
    "    )\n",
    "    # 19 < age ≤ 40\n",
    "    phenotype_df = phenotype_df.withColumn(\n",
    "        f\"{name}_incident_mid2\",\n",
    "        F.when(\n",
    "            (F.col(\"diagnosis_age\") > 19) & (F.col(\"diagnosis_age\") <= 40), F.lit(1)\n",
    "        ).otherwise(F.lit(0)),\n",
    "    )\n",
    "    # 40 < age ≤ 60\n",
    "    phenotype_df = phenotype_df.withColumn(\n",
    "        f\"{name}_incident_meno2\",\n",
    "        F.when(\n",
    "            (F.col(\"diagnosis_age\") > 40)  & (F.col(\"diagnosis_age\") <= 60), F.lit(1)\n",
    "        ).otherwise(F.lit(0)),\n",
    "    )\n",
    "    # over 60\n",
    "    phenotype_df = phenotype_df.withColumn(\n",
    "        f\"{name}_over60\",\n",
    "        F.when(F.col(\"diagnosis_age\") > 60, F.lit(1)).otherwise(F.lit(0)),\n",
    "    )\n",
    "    # ever variable: filter out over 60s\n",
    "    phenotype_df = phenotype_df.withColumn(\n",
    "        f\"{name}_ever\",\n",
    "        F.when(F.col(f\"{name}_over60\") == 1, F.lit(0)).otherwise(F.lit(1)),\n",
    "    )\n",
    "\n",
    "\n",
    "    # calculate summary statistics\n",
    "    summary_stats = (\n",
    "        phenotype_df.agg(\n",
    "            F.sum(F.col(f\"{name}_ever\")).alias(\"n\"),\n",
    "            F.avg(\n",
    "                F.when((F.col(\"diagnosis_age\").isNotNull()) & (F.col(f\"{name}_ever\") == 1), F.col(\"diagnosis_age\"))\n",
    "            ).alias(\"avg_diagnosis_age\"),\n",
    "            F.sum(\n",
    "                F.when(F.col(\"diagnosis_age\").isNull(), 1).otherwise(0)\n",
    "            ).alias(\"null_ages\"),\n",
    "            F.sum(F.col(f\"{name}_incident_ado\")).alias(f\"{name}_incident_ado_count\"),\n",
    "            F.sum(F.col(f\"{name}_incident_mid2\")).alias(f\"{name}_incident_mid_count2\"),\n",
    "            F.sum(F.col(f\"{name}_incident_meno2\")).alias(f\"{name}_incident_meno_count2\"),\n",
    "            F.sum(F.col(f\"{name}_over60\")).alias(f\"{name}_over60_count\"),\n",
    "        )\n",
    "        .collect()[0]\n",
    "        .asDict()\n",
    "    )\n",
    "\n",
    "    summary_stats[\"name\"] = name\n",
    "\n",
    "    return phenotype_df, summary_stats\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4d081900-29f0-482f-9bc1-0e364b74ab33",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def print_summary_statistics(counts, n=None):\n",
    "    if not n:\n",
    "        n = counts[\"n\"]\n",
    "    \n",
    "    name = counts[\"name\"]\n",
    "    percent = lambda x: (x / n) * 100\n",
    "    print(f\"---------------------------- {name} summary statistics ----------------------------\")\n",
    "\n",
    "    categories = [\n",
    "        (\"age ≤ 19\", counts[f\"{name}_incident_ado_count\"]),\n",
    "        (\"19 < age ≤ 40\", counts[f\"{name}_incident_mid_count2\"]),\n",
    "        (\"40 < age ≤ 60\", counts[f\"{name}_incident_meno_count2\"]),\n",
    "        (\"age unavailable\", counts[\"null_ages\"])\n",
    "    ]\n",
    "\n",
    "    print(f\"{'Age Category':<20} {'Count':<10} {'Percentage':<12}\")\n",
    "    print(\"-\" * 50)\n",
    "    \n",
    "    for category, count in categories:\n",
    "        print(f\"{category:<20} {count:<10} {percent(count):.2f}%\")\n",
    "        \n",
    "    print(\"--------------------------------------------------------------------------------\")\n",
    "    print(f\"{name} diagnoses over 60 (excluded): {counts[f'{name}_over60_count']}\")  \n",
    "    print(f\"total women with {name}: {n}\")\n",
    "    print(f\"average diagnosis age: {counts['avg_diagnosis_age']:.2f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c65536ed-a866-4381-b88a-fa58c4a7742a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# plot age categorisation pie chart\n",
    "def plot_counts_pie_chart(counts):\n",
    "    name = counts[\"name\"]\n",
    "    percent = lambda x: (x / counts[\"n\"]) * 100\n",
    "\n",
    "    labels = [\"age ≤ 19\", \"19 < age ≤ 40\", \"40 < age ≤ 60\",  \"age unavailable\"]\n",
    "    frequencies = [\n",
    "        counts[f'{name}_incident_ado_count'],\n",
    "        counts[f'{name}_incident_mid_count2'],\n",
    "        counts[f'{name}_incident_meno_count2'],\n",
    "        counts['null_ages']\n",
    "    ]\n",
    "\n",
    "    colors = ['gold', 'lightblue', 'lightcoral', 'limegreen'] \n",
    "\n",
    "    plt.figure(figsize=(8, 6))\n",
    "    wedges, texts = plt.pie(\n",
    "        frequencies,\n",
    "        labels=None,        \n",
    "        colors=colors,\n",
    "        startangle=140,\n",
    "        wedgeprops={'edgecolor': 'black'}  \n",
    "    )\n",
    "\n",
    "    plt.legend(\n",
    "        labels=[f\"{label} ({percent(frequencies):.2f}%)\" for label, frequencies in zip(labels, frequencies)], \n",
    "        title=\"Age Groups\", \n",
    "        loc=\"center left\", \n",
    "        bbox_to_anchor=(1, 0.5), \n",
    "        fontsize=12,              \n",
    "        title_fontsize=14        \n",
    "    )\n",
    "\n",
    "    plt.title(f\"Age distribution for {name}\", fontsize=18)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "cd0a42af-eebd-4087-94cc-68c13409e99e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "hmb_phenotype_df_, hmb_counts = derive_variables(hmb_phenotype_df, \"hmb\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "5821bc0e-e7fb-47e5-b272-2cb7b4a2bf44",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---------------------------- hmb summary statistics ----------------------------\n",
      "Age Category         Count      Percentage  \n",
      "--------------------------------------------------\n",
      "age ≤ 19             72         0.24%\n",
      "19 < age ≤ 40        6536       22.04%\n",
      "40 < age ≤ 60        22704      76.56%\n",
      "age unavailable      343        1.16%\n",
      "--------------------------------------------------------------------------------\n",
      "hmb diagnoses over 60 (excluded): 114\n",
      "total women with hmb: 29655\n",
      "average diagnosis age: 44.69\n",
      "% of women with hmb: 10.86%\n"
     ]
    }
   ],
   "source": [
    "plot_counts_pie_chart(hmb_counts)\n",
    "print_summary_statistics(hmb_counts)\n",
    "print(f\"% of women with hmb: {((hmb_counts['n'] / 273157) * 100):.2f}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb020d24-461c-4b45-b959-fc2e14e51506",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Control group"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f962eed-a1bb-45ec-a53c-82cfdc97f1c3",
   "metadata": {},
   "source": [
    "Control: Any females with EHR data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7bfe6b23-08e2-4f40-95c0-e18cd3b0656c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "245942"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filter female eids from all EHR tables\n",
    "filter_females = lambda df: females_df.join(df, on=\"eid\", how=\"inner\")\n",
    "\n",
    "# females with icd9 or icd10 codes\n",
    "icd_df = filter_females(ukb.spark.sql(f\"SELECT DISTINCT eid FROM `{ukb.database}`.`hesin_diag` WHERE diag_icd9 IS NOT NULL OR diag_icd10 IS NOT NULL\"))\n",
    "icd_df.count() # 245942"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0a8e1d1a-626b-40fa-8c4a-f6719304ba8e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "125700"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# females with gp clinical data\n",
    "gp_df = filter_females(ukb.spark.sql(f\"SELECT DISTINCT eid FROM `{ukb.database}`.`gp_clinical` WHERE read_2 IS NOT NULL OR read_3 IS NOT NULL\"))\n",
    "gp_df.count() # 125700"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ff4c80c6-f098-47dd-b5a4-75470078f5a5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "18191"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# females with death records\n",
    "death_df = filter_females(ukb.spark.sql(f\"SELECT DISTINCT eid FROM `{ukb.database}`.`death_cause` WHERE cause_icd10 IS NOT NULL\"))\n",
    "death_df.count() # 18191"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b9ccf4c7-80d6-42c9-9fcb-ef5187cc55e2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "245197"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# females with OPCS codes\n",
    "opcs_df = filter_females(ukb.spark.sql(f\"SELECT DISTINCT eid FROM `{ukb.database}`.`hesin_oper` WHERE oper3 IS NOT NULL OR oper4 IS NOT NULL\"))\n",
    "opcs_df.count() # 245197"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8b22a91d-d1a1-44fa-9e85-d2eaf45fe77c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "63447"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# females with cancer records\n",
    "case_statements = [f\"p40006_i{i} IS NOT NULL\" for i in range(22)]\n",
    "case_condition = \" OR \".join(case_statements)\n",
    "query = f\"SELECT DISTINCT eid FROM `{ukb.database}`.`participant_0073` WHERE {case_condition}\"\n",
    "cancer_df = filter_females(ukb.spark.sql(query))\n",
    "cancer_df.count() # 63447"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "d3806e57-a97f-42db-a3c2-1ae1b248d180",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "258798"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# join all EHR recorded females to make control DF\n",
    "control_df = icd_df.join(gp_df, on=\"eid\", how=\"outer\")\n",
    "control_df = control_df.join(death_df, on=\"eid\", how=\"outer\")\n",
    "control_df = control_df.join(opcs_df, on=\"eid\", how=\"outer\")\n",
    "control_df = control_df.join(cancer_df, on=\"eid\", how=\"outer\")\n",
    "control_df.count() # 258798"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80367daa-b0fe-44a9-8a52-81ffacba5c17",
   "metadata": {},
   "source": [
    "## HMB phenotype with control"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "b68706ed-ea36-4010-b859-2e3092b53541",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "258908"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hmb_final_df = control_df.join(hmb_phenotype_df_, on=\"eid\", how=\"outer\")\n",
    "hmb_final_df = hmb_final_df.withColumn(\"hmb_ever\", F.when(F.col(\"hmb_ever\").isNull(), F.lit(0)).otherwise(F.col(\"hmb_ever\")))\n",
    "hmb_final_df = hmb_final_df.withColumnRenamed(\"diagnosis_age\", \"hmb_diagnosis_age\")\n",
    "hmb_final_df = hmb_final_df.select(*[\"eid\", \"hmb_ever\", \"hmb_diagnosis_age\", \"hmb_incident_ado\", \"hmb_incident_mid2\", \"hmb_incident_meno2\", \"hmb_over60\"])\n",
    "hmb_final_df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96ab8e04",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Menstrual Pain (MP) phenotype"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e13cd1b1-8868-405d-9d73-b582ef5e7ebe",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Self-report questionnaire"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d2f657b-67ee-4178-8e48-06505a2a8689",
   "metadata": {},
   "source": [
    "- p21050 found in participant_0042             \n",
    "- p21026 found in participant_0042        \n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3687cdb-251e-4f34-95ee-7b2888f77add",
   "metadata": {},
   "source": [
    "Self-report questionnaire (available for MP only - not HMB)         \n",
    "• 21050: During the past 3 months, how much have you been bothered by menstrual cramps or other problems with your periods? [n bothered a lot =1083]            \n",
    "-600\tNot bothered at all       \n",
    "-601\tBothered a little       \n",
    "-602\tBothered a lot         \n",
    "-818\tPrefer not to answer         \n",
    "-313\tNot applicable         \n",
    "\n",
    "• 21026: Did this discomfort or pain occur only during your menstrual bleeding and not at other times? [n yes =1121]          \n",
    "1\tYes    \n",
    "0\tNo     \n",
    "-818\tPrefer not to answer     \n",
    "-313\tNot applicable         \n",
    "\n",
    "Participants answering “bothered a lot” to 21050 and “yes” to 21026 should be coded as a case!          "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "72294fe8-e8fa-4b91-9e60-29280399e6ba",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# menstrual pain self report variable\n",
    "query = f\"SELECT eid, p21050, p21026 FROM `{ukb.database}`.`participant_0042` WHERE p21050 == -602 AND p21026 == 1\"\n",
    "mp_selfrep_df = ukb.spark.sql(query)\n",
    "mp_selfrep_df = mp_selfrep_df.select(\"eid\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82e5ea34-a4c7-4158-992d-4f8093022d0a",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Mefenamic acid"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8497bd9-ce0e-4dd7-b34d-e614f9186610",
   "metadata": {},
   "source": [
    "Mefenamic acid code = 1140871542"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "5b8f7ec6-be12-451c-b2f0-eeb7976c2c99",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# all fields potentially containing a treatment code\n",
    "columns = [f\"p20003_i{i}_a{a}\" for i in range(4) for a in range(48)]\n",
    "\n",
    "# extract eids with treatment code for mefenamic acid\n",
    "query = f\"SELECT DISTINCT eid FROM `{ukb.database}`.`participant_0032` WHERE {' OR '.join([f'{col} = 1140871542' for col in columns])}\"\n",
    "\n",
    "mefenamic_df = ukb.spark.sql(query)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48776694-140a-4ca7-83fb-1f51850f1f86",
   "metadata": {},
   "source": [
    "## ICD0, ICD10, Read2 and CTV3 Codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "60e3d867-6dde-4dcf-84f6-d4ed63e23c24",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 1 ICD9 codes...\n",
      "Operation complete: Found 313 matching records in 8.86 seconds.\n"
     ]
    }
   ],
   "source": [
    "icd9 = [\"6253\"]  # dysmenorrhoea\n",
    "mp_icd9 = ukb.get_icd9(icd9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "57e6314d-e7b7-4999-a1fc-000b168e7ffa",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "ICD10 = [\n",
    "    \"N94.4\",  # primary dysmenorrhoea\n",
    "    \"N94.5\",  # secondary dysmenorrhoea\n",
    "    \"N94.6\"   # dysmenorrhoea, unspecified\n",
    "]\n",
    "mp_icd10_hesin = ukb.get_icd10_hesin(ICD10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "4de2e86b-898f-4d7f-b195-527f5cd5264a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "Read2 = [\n",
    "    \"Eu45y11\",  # psychogenic dysmenorrhoea\n",
    "    \"K583.00\",  # dysmenorrhoea\n",
    "    \"1574.00\",  # h/o: dysmenorrhoea\n",
    "    \"1574.11\",  # h/o: painful periods\n",
    "    \"K583.11\",  # painful menorrhoea\n",
    "    \"K583.12\",  # painful menstruation\n",
    "    \"K583.13\",  # period pains\n",
    "    \"K583000\",  # primary dysmenorrhoea\n",
    "    \"E265200\",  # psychogenic dysmenorrhea\n",
    "    \"K583100\",  # secondary dysmenorrhoea\n",
    "    \"K583.14\"   # spasmodic dysmenorrhoea\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "c46afcc5-61f2-48aa-876c-1e53e2b3b673",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 11 Read2/CTV3 codes...\n",
      "Operation complete: Found 1510 matching records in 5.46 seconds.\n"
     ]
    }
   ],
   "source": [
    "mp_gp_clinical = ukb.get_gp_clinical(Read2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b783bdf-0b47-40c8-ab21-939c2aec3a49",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Final MP phenotype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "ad14731c-5994-42ee-bcc5-e2c8feb831d2",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# join all dates entries into one table\n",
    "all_dates_df = mp_icd9.join(mp_icd10_hesin.withColumnRenamed(\"diagnosis_date\", \"icd10_diagnosis_date\"), on=\"eid\", how=\"outer\")\n",
    "all_dates_df = all_dates_df.join(mp_gp_clinical, on=\"eid\", how=\"outer\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "d7decf57-24f6-4e24-b870-b6c9373d2581",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# get the earliest date across all fields\n",
    "first_date_df = all_dates_df.withColumn('first_diagnosis_date', F.least(F.col('diagnosis_date'), F.col('icd10_diagnosis_date'), F.col('event_dt')))\n",
    "first_date_df = first_date_df.select(\"eid\", F.year(\"first_diagnosis_date\").alias(\"first_diagnosis_date\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "147bf2ce-00c8-42c5-b186-08fc5dc152e8",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# join dobs with first diagnosis dates\n",
    "dates_to_ages_df = first_date_df.join(dob_df, on=\"eid\", how=\"inner\")\n",
    "\n",
    "# subtract date of birth from diagnosis date to get age at diagnosis\n",
    "dates_to_ages_df = dates_to_ages_df.withColumn(\n",
    "    \"first_diagnosis_age_unfiltered\", \n",
    "    dates_to_ages_df[\"first_diagnosis_date\"] - dates_to_ages_df[\"p34\"]\n",
    ")\n",
    "\n",
    "# replace ages <= 0 with null in a column \n",
    "dates_to_ages_df = dates_to_ages_df.withColumn(\n",
    "    'diagnosis_age',\n",
    "    F.when(F.col('first_diagnosis_age_unfiltered') <= 0, F.lit(None))  # if value <= 0, set it to null\n",
    "    .otherwise(F.col('first_diagnosis_age_unfiltered'))                # otherwise keep the original value\n",
    ")\n",
    "\n",
    "all_ages_df = dates_to_ages_df.select(\"eid\", \"diagnosis_age\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "2cbd9684-2706-4f02-9b03-77be15c992a5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# finally join mefenamic acid and self reoport participants (for whom no dates available)\n",
    "mp_phenotype_df = all_ages_df.join(mefenamic_df, on=\"eid\", how=\"outer\")\n",
    "mp_phenotype_df = mp_phenotype_df.join(mp_selfrep_df, on=\"eid\", how=\"outer\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b89b116-843a-456f-a862-1b15f206b4db",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Filtering female participants only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "0c640970-bbaa-4810-b687-9fc4495566a8",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "mp_phenotype_df = females_df.join(mp_phenotype_df, on=\"eid\", how=\"inner\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96d2d32e-c2d9-4947-b08a-14c6e7134158",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Deriving MP variables"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14c62faa-ad00-4d59-969e-461e27deb0d7",
   "metadata": {},
   "source": [
    "Derived variables         \n",
    "• mp_ever <- any diagnosis ever recorded vs no diagnosis ever recorded        \n",
    "• mp_incident_ado <- any diagnosis before or equal to age 19 vs no diagnosis before or equal to age 19         \n",
    "• mp_incident_mid <- any diagnosis between age 20-44 (inclusive) vs no diagnosis between age 20-44 (inclusive)      \n",
    "• mp_incident_meno <- any diagnosis between age 45-60 (inclusive) vs no diagnosis between age 45-60 (inclusive)      \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "f281d937-2c9e-4e9b-b205-de7597f950dd",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "mp_phenotype_df_, mp_counts = derive_variables(mp_phenotype_df, \"mp\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "6e82a497-3572-4cf0-b3a6-b67feae5fa27",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvsAAAH9CAYAAACEMLPOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAADHNUlEQVR4nOzddXhT5/vH8XdSF+rQ0tJCgeI23F3GsOFs+NDhDGcbDsN9AzYYMmDYcDZcxoCx4dpCobi2SAtNajm/P/g2v5Uq0Pa06f26rlwbJ0c+Jw3lzpNHNIqiKAghhBBCCCFMjlbtAEIIIYQQQoi0IcW+EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhImSYl8IIYQQQggTJcW+yDAOHz6MRqNBo9HEe27FihVoNBry5MmT/sFS4NatW8bst27divNcUveVkXTp0gWNRkOXLl3UjvJeli9fTqVKlXBwcDC+3nPnzlU7lhBCCKEqc7UDiIQ9f/4cT09P9Ho9ANeuXcPPz0/lVKbj3LlzbN26FScnJwYNGqR2nDR1+PBhDh8+TJ48eTJtIZ+cWbNmMXToUADMzc3JkSMHGo0GOzs7lZMJIYQQ6pKW/QxqzZo1xkIf4Oeff1YxjfocHR0pWLAg+fLlS5XznTt3jvHjx6day6+FhQUFCxakYMGCWFhYpMo5U8vhw4cZP348K1asSHK/nDlzUrBgQXLmzJk+wVLRjBkzABgwYADh4eE8fvyYR48e0aNHD5WTCSGEEOqSYj+DWrZsGQD9+/cHYOXKlcTExKgZSVXNmzfH39+fAwcOqB0lQV5eXvj7++Pv74+Xl5facd7Ld999h7+/P999953aUd7J06dPefz4MQA9evTIcB+2hBBCCDVJsZ8BnTlzhnPnzuHk5MT06dPx9fXl4cOH/P7772pHEyLDCQ8PN/6/vb29ikmEEEKIjEeK/QwotlW/bdu2WFtb06lTJyDlXXm2bdtG7dq1cXJywt7enpIlSzJ9+nSioqIYN24cGo2GmjVrJnr8rVu3GDRoEEWLFsXe3h5bW1sKFSrEwIEDuXPnzgfdm7+/P+3bt8fDwwNra2vy5s1L//79jS2ziUlugO7Jkydp3749vr6+WFtbY2dnR+7cualRowYTJ07k3r17xn01Gg1du3YF4Pbt28bBnLGPcePGGff976BVRVFYunQpVatWxdXVFY1GY+wak9QA3bedOnWKVq1akTNnTqytrcmfPz/Dhg3jxYsXCe6fkp9ZQoOAYzONHz8egCNHjsS71/927UnJAN3Dhw/TunVrvLy8sLKyws3NjTp16rB8+fJEv3l6O/+BAwdo1KgR2bNnx9ramsKFCzN+/Pg43dZSIvae//ue8PX1Nd5bQu+V1Mj/22+/Ub9+fXLkyIFWq43zfknK2++R27dv06NHD3x8fLC2tiZfvnx88803vH792njMpUuX6NChA97e3lhbW+Pn58ekSZOIiopK8Bo1a9Y0vocjIyOZOnUqJUqUwM7ODmdnZ+rVq8cff/yRorxCCCFMhCIyFJ1Opzg5OSmAcuzYMUVRFOXGjRuKRqNRzM3NlUePHiV5/JAhQxTA+HByclLMzc0VQKlevboyevRoBVBq1KiR4PGrV69WrKysjMdbWVkpNjY2xj9ny5ZN2bNnz3vd2x9//BHn3Pb29oq1tbUCKDlz5lR+/vln43NvW758uQIouXPnjvfcihUrFI1GEyezg4NDnNdh+fLlxv3d3d2Nz2u1WsXd3T3OY8aMGcZ9O3furABKp06dlJYtWxqPcXZ2VrRarfG8QUFBxmsFBQXFyXfo0CHjc1u3blUsLS0VQHFwcDD+f+y9vX2soijK2LFjk/yZvX2NWHfu3FHc3d0VOzs7BVAsLCzi3eu6devi3Wvnzp0TvMbgwYON19BoNIqTk5NiZmZm3Fa7dm0lNDQ0yfzTp09XNBqN8fj//txq1aqlREdHJ3qPbzt27Jji7u6uuLm5Gc/h5uZmvLeyZcumev6vvvrKeLyzs7NiZmamjB07NkV5//se+e2334x/zx0cHOLkqFatmhIZGans3LlTsbW1VQDF0dExzmvVtm3bBK9Ro0YNBVBGjRqlVKtWTQEUc3Nz47ViHynNLIQQIvOTYj+DWb16tQIo+fPnj7M99h/u6dOnJ3rsr7/+avzH/PPPP1fu3bunKMqbDxA//vijYm1trTg7OydaOO7du1fRarWKubm5Mnz4cCUoKEgxGAyKwWBQ/P39ldatWxuLk9u3b7/Tfd29e9dYYJcoUUI5efKkoiiKEhMTo/zxxx9Krly54hQkb0us2H/9+rWSLVs2BVA6dOigBAYGGp979eqVcurUKWXYsGHKrl27UnS+t8UWwPb29oq5ubkyc+ZM5eXLl4qiKEpYWJjy4MEDRVFSXuw7OjoqNWvWVK5cuaIoiqJERUUp69evN/5cypUrF6/gfd9i/12O/++9JlTsL1iwwHj+nj17Kg8fPlQU5c1rPGfOHOMHyoSK0NjrOzk5KVqtVhk1apTy9OlTRVEU5eXLl8qYMWOM5162bFmSGROS1Gufmvnt7e0VQBkxYoTy5MkTRVEURa/XK7du3XrnnE5OTkqdOnWUy5cvK4qiKOHh4cr8+fONRf8333yjODo6Km3btjWePywsTPn666+N59i3b1+8a8QW+46OjoqVlZWyePFiRafTKYry5sNfq1atjMdv27YtRbmFEEJkblLsZzC1atVSAGXChAlxtv/0008KoBQqVCjB4wwGg5I/f34FUOrVq6cYDIZ4+8QWuAkVfjExMYqfn58CKEuWLEk0X9OmTRVAGThw4Dvd15dffqkAiqurq/L48eN4z1+8eFGxsLB452L/5MmTCqDY2dkpUVFRKc7zrsU+oMyfPz/R/VJa7BcoUEAJDw+Pd/y+ffuM+2zYsCHOc2oX++Hh4YqLi4sCKJ999lmCx86fP994/VOnTiV4/aRalFu0aKEASt26dZPMmJDkiv3UzP/VV1+9c76EchYtWlTR6/Xx9unYsaNxn8T+Hsd+8O/WrVu852KL/cQ+OMXExCjVq1c3ZhBCCGH6pM9+BnLz5k1jP+SOHTvGea5NmzbY2Njg7+/P8ePH4x177tw5AgMDARg9enSCCzh17twZHx+fBK/9559/cv36ddzc3OjevXuiGWPHD+zZsyfF96UoCuvXrwegd+/e5MiRI94+xYoVo1WrVik+ZywnJycAIiMjCQkJeefjU8rZ2ZlevXp98HmGDRuGjY1NvO1169alcuXKAKxbt+6Dr5Oa9u3bx7NnzwAS7Z/ep08f45Sda9euTXAfKysr41z4b2vWrBkAFy5c+MC08aVWfq1Wy4gRI1Il0+DBg7Gysoq3vUGDBsb/HzlyZIJ/j2P3Seq18vb2No5L+S+tVss333wDwOXLl7l48eI7ZxdCCJG5SLGfgSxfvhxFUahWrVq8wYUODg58+umnwP8P4P2vM2fOAG/me48tGt+m0WioUaNGgs8dO3YMgJcvX+Lp6YmHh0eCj9h5y2/fvp3i+woKCjIWW7Vr1050v6SeS0y+fPkoVKgQUVFRVKhQgWnTpnHu3LlUn6a0XLlyWFpafvB5UnL/p06d+uDrpKbYPN7e3hQoUCDBfczMzJLNHzvgOyGenp4AxvdJakqt/Pnz50/wg+r7KF++fILb3d3djf9frly5JPd5/vx5ouePHaibkGrVqmFu/mY9xYz2XhNCCJH6pNjPIAwGg3FmlNjW87d17twZgA0bNvDq1as4zz19+hQAV1fXJIvSxOaAf/DgAQBRUVE8fvw40UdsgaHT6VJ8b0+ePEn2+gC5cuVK8TljmZmZsW7dOnx9fbl9+zYjR47ko48+wsHBgXr16rFo0aI4UzO+r9Qq8pK6/9jn/vt6ZQSxeZJbPyD255dY/mzZsiV6bGzxGR0d/T4Rk5Ra+VPrPQCJvxaxr0NK9klsRh5I+l6tra1xdXUFMt57TQghROqTYj+D2LNnj3F6yO7du8ebIlGj0fDxxx8D8OrVKzZs2JDgeRJrzUtObEt4hQoVUN6M5Uj2kVGULFkSf39/fvvtN3r27EmxYsXQ6XTs37+fPn36UKhQoQ/urmBmZpZKaUVmJe8BIYQQmZEU+xlEQl1z3mX/7NmzAxAcHExkZGSix92/fz/B7R4eHsC7dc9Jqf+2iCZ2/eSeS46lpSUtWrRgyZIlXLx4kadPn7J48WJcXFy4e/eu8VsRtaXk/t9uQY5tyU1qHvqXL1+mQrqExeb571oFCYl9PjVbwFNDZs//PpJ6n0VERBjHt5jCvQohhEiaFPsZwNOnT9m+fTsAmzZtIiwsLNHHP//8A8Dx48cJCAgwnqN06dLAm6/2ExrAC28Gyv75558JPlelShUAHj16lOr9eH19fXFxcQHg0KFDie538ODBVLumq6srvXr1Ytq0aQCcPXs2zgBerfbNWz+9v6FI6v5jnytbtmyc7c7OzgDcvXs30WNPnjyZ6HMfeq+xee7du8e1a9cS3CcmJsaYP7G+5mrJ7Pnfx5EjRxL9eR89etTYXert95oQQgjTI8V+BvDLL78QFRWFo6MjTZo0wd7ePtFHuXLlKFSoEBC3db9UqVLkz58fgKlTpyb4D/3q1asTbbmvVauW8fjBgwcn+e0AvNtASo1GQ5s2bQBYvHgxwcHB8fa5cuUKmzZtSvE5Y0VERCT5/H9nvokteuHNgGcg0VVr08rMmTMTbKE/dOiQcZB027Zt4zxXsmRJ4M24ioSK+idPnvDTTz8les0Pvdd69eoZ+3gnNpvNkiVLjOM+Pvvss/e6TlrJ7Pnfx507d1i5cmW87QaDgSlTpgBQpEgRihcvnt7RhBBCpDMp9jOA2KK9WbNmKZrxpXXr1gCsWrXK2EKn0WgYP3488Kb/f+fOnY3Fi16vZ9myZfTq1cvYSvw2c3NzFi9ejLm5OX/99RfVq1fnwIEDcQYB3rx5k8WLF1OuXDl++OGHd7rHUaNGkS1bNoKDg6lXr57x2wNFUdi7dy8NGzbE1tb2nc4Jb6aprFKlCkuWLOHmzZvG7TExMezZs4eRI0cCUKlSpTj3XqxYMQBCQ0MTHf+QFh4+fEijRo2M38pER0ezadMm47SjpUuXpkWLFnGOqVy5Mrlz5wbeDNI+deoUiqJgMBg4fPgwNWvWxGAwJHrN2Hu9fPlyot/6JMXGxsZYJP/666/07t2bx48fAxAeHs78+fMZNGgQ8OaDSpkyZd75Gmkps+d/H46Ojnz55Zf89NNPxg+Xd+/e5bPPPjN+gzFp0iQ1IwohhEgv6Tyvv3jLiRMnjIvg7NixI0XHXLhwwXjM1q1b4zw3aNAg43MajUZxdnY2LlZVu3ZtZdSoUQqgNGjQIMFzb9myxbgiLaBYWFgorq6uipWVlXEboEyaNOmd73Xnzp1xzpMtWzbFxsZGAZScOXMqP//88zsvqvXfhcIAxcrKSnF1dVW0Wq1xm6enp3L16tV456xTp06cLLlz51Zy586tzJkzx7hPUqvK/ldKF9XaunWr8ecRu8pp7HM+Pj7KzZs3Ezz/7t274yw6Zmtrq1hbWyuA4ufnF2f15LdFRUUpBQsWND7v7OxsvNeNGzem+F4HDx4c770Vu/IsoNSqVUsJDQ2Nd9yHLgqWnJSsoJvW+VMrZ0peh6QWhItdVGvUqFFK1apVjX+HY1dojn188803H3QvQgghMg9p2VdZbKu+o6Mj9evXT9ExxYsXp3DhwnGOjzVnzhw2b95MzZo1yZYtGxERERQuXJgZM2awZ88eXr9+Dfz/YlRv+/TTTwkMDGTs2LGUL18ee3t7Xrx4gZWVFSVLlqR79+5s2bKFYcOGvfO9NmrUiDNnztCuXTty5MhBZGQk7u7u9OvXj7Nnz+Lr6/vO52zatCmrVq2ia9eulCxZEkdHR16+fEm2bNkoX748EydO5PLly8auT/+1adMmBg8eTIECBYiKiuL27dvcvn07Tbv2NGvWjOPHj9OyZUusra1RFAVfX1+GDBnCuXPnEn0NGjRowNGjR2ncuDHOzs7ExMTg7e3NyJEjOX36tHGAdULMzc05cOAA3bt3x9fXl9evXxvv9e0pXJMye/ZsDh48SMuWLXF3d+fVq1dky5aNWrVq8fPPP7Nv374kp9dUW2bP/y4sLS05cOAAU6ZMoWDBgkRERODo6EidOnXYtWsXEydOVDuiEEKIdKJRlAw0h6JIc1WqVOH48eNMmDCBb7/9Vu04QohUVLNmTY4cOcLYsWMTHZ8ghBAia5GW/SzkyJEjxj7bsXP2CyGEEEII0yXFvonp27cvK1as4NGjR8YZeV68eMGSJUto1qwZALVr1zaJ6QWFEEIIIUTSzJPfRWQmx44dM86UY2Vlha2tLS9evDAW/kWKFGHVqlVqRhRCCCGEEOlEin0TM2HCBLZu3crJkyd5/PgxL1++xNnZmaJFi9KiRQt69uz5XlNcCiGEEEKIzEcG6AohhBBCCGGipM++EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhImSYl8IIYQQQggTJcW+EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhImSYl8IIYQQQggTJcW+EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhImSYl8IIYQQQggTJcW+EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhImSYl8IIYQQQggTJcW+EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhImSYl8IIYQQQggTJcW+EEIIIYQQJkqKfSGEEEIIIUyUFPtCCCGEEEKYKCn2hRBCCCGEMFFS7AshhBBCCGGipNgXQgghhBDCREmxL4QQQgghhIkyVzuAEOINRVEICwvj8ePHPH36FL1eT2RkZLxHREREvG1RUVFYW1tjZ2eHra1tsv+1t7fHyspK7VsWQgghRBqTYl+INBYZGUlQUBAPHz7k8ePHcR6PHj/iwaMHPH78mJCnIUToI5I/oQbMLM3QWmgxszRDY6FBY6bBEGkgRhdDtC4axaAkexp7R3u8vLzw9vLGO5c3Xl5e5MqVCy8vL+PDzc0NrVa+ABRCCCEyK42iKMlXBUKIZD179gx/f3/j46r/VS5fvcydoDvExMQY9zOzNMPK1QpzF3M0zhrMXc0xdzHHwsUCc9f//dfZHK2V9k0hb6lBY6FBa6598/9mmiRzKIqCEqlg0Bsw6Azx/hujj8Hw2kBUSBRRT6KIehqF4YmBqKdR6J/q43xQsLC0wD2nO965vClcsDBFihShcOE3//Xx8ZEPAkIIIUQGJ8W+EO8oJCSEf/75hytXruDv78/lq5fxD/DnefDzNztowNbLFovcFljlscI6jzVWPlZYZLfAwsUCrb0WjSbpgl0tSrRC1LP/fQh4EkXkk8g3/30cSfSdaHRBOqJ10QBY21rj7OJMreq1KFu2LCVLlqREiRK4ubmpfBdCCCGEiCXFvhBJiImJ4fLly5w4cYITJ05w9PhRbl6/CYC5rTk2uW0wz22OdR5r48PK2wqttWm2eCsGhcjHkehv6gk9HsrTdU/J7eLC47Aw9FFRAHh7eVGpShUqV65M5cqVKVWqFBYWFionF0IIIbImKfaF+I/nz5/z999/c/z4cY6dOMY///zD67DXaMw02Bewx7q4NXbF7bArYYell2WGbaFPDyHbQ7g9/jbXhw7FxdaWGyEhXH78mLMPHvDvgwecuX+fiKgobKytKVe2LJWrVqVy5cpUqlRJWv+FEEKIdCLFvsjSXr16xYEDB/j99985eOQggQGBAFg5W2FT3Abb4rbYlbTDtogtZjZmKqfNWO7NvsezjSEEj/42wecjo6O58OgRJ+/e5Z+7d/nn/n0evnwJQEE/Pxo0bEiDBg2oWbMmtra26RldCCGEyDKk2BdZTmBgILt27WLHzh38+eefREVGYZfHDpsyNtiVfNNqb5XLKku32qfE9T7XsbmscG3o0BTtrygKd1++5N+7d/nr1i0OBAVx59kzrCwtqVatGh//r/gvWrSovPZCCCFEKpFiX5i8yMhIjh49yq5du9i2cxs3r99Ea6ElW5lsZKuSDYdqDlh7W6sdM9O5UPcCZe1zsrtbt/c6XlEUAkNC2B8YyMGbNzkaFIQ+KgqvnDlp0LAhn3zyCQ0bNpRWfyGEEOIDSLEvTFJwcDA7duxgx84d7N23l9dhr7HOYY1dZTscqzqSrUI2zGylW877in4RzYU6F+hdoQJTGzZMlXPqoqI4cfs2B27c4MDNm/g/foytjQ2NGzemTdu2UvgLIYQQ70GKfWEyXr9+zfbt21m9ZjV79uwhJiYGh+IO2FW1w7GKIzYFbaR7SCoJOx3G9Z7XWduuHZ8UKpQm17gZEsLWK1fYcvUqFx88wM7WlsZNmtC6dWsp/IUQQogUkmJfZGrR0dHs27eP1WtWs2XrFnSvdTiUcMCxoSNOdZ2wcJEpH9PC0w1PuTv9Lg9GjcbW0jLNr3cjJIStly+z9epVLj58aCz8P//8cz755BPMzWUxcCGEECIhUuyLTOnq1assX76cFatW8PTxU+zy2uHwsQMuDVywymWldjyTd2fKHcJ2veDxyG/S/dqBwcHGFv/LDx+S092dLl98Qbdu3ciXL1+65xFCCCEyMin2Rabx8uVL1q1bx9Kfl3Lqn1NYOlri+LEjrk1csSkkXXTSU0CXAFzvmHNh8GBVc5x78IDVZ8+y8dIlXup01K5Zk+49e9K8eXOsrWXQtRBCCCHFvsjwrl+/zrx58/h5xc/odXocKzvi3MQZx+qOaC1Nc6XajExRFM5XO09NT182d+yodhwAwiMj2XblCqvOnePErVu4ODnRsXNnunfvTrFixdSOJ4QQQqhGin2RISmKwuHDh5k1exa/7/odS2dLnFs649bSDcvsad9HXCQu8lEklxpdYlj16nxdu7baceK5HhzML2fOsPbCBYJfvaJmjRoMGTqUTz75BK1WPhwKIYTIWqTYFxlKREQE69atY9acWVw8fxG7fHa4fu6KS0MXtFZSqGUEL4+95MaAG/zetSuVc+dWO06iIqOj2eXvz/cnT3Lq7l0KFSjAV0OH0rFjR+niI4QQIsuQYl9kCMHBwSxevJj5C+fz9PFTnKo44fa5G9kqZJO++BnMo5WPeLjwAcHfjMEsE7SUK4rCybt3WXjiBLv8/XFzcaFv//706dOH7Nmzqx1PCCGESFNS7AtVXb16ldmzZ7Nq9SpilBicGzuT/bPs2PjaqB1NJOLWmFtEHHzF/RGj1Y7yzm6EhLDo779Zc/48ikZD5y5dGDJkCH5+fmpHE0IIIdKEFPtCFTdu3GDsuLGsXbMWKzcrnFs7k71FdsydZb70jO5q26vkem7Lv/37qx3lvT0LD2fZv//y06lTBL9+TceOHRkzZgx58+ZVO5oQQgiRqqTYF+nq/v37TJw4kaXLlmLhbEH2L7Lj+qmrzKqTSSjRCueqnKNx/oL80q6d2nE+mD4qipVnzjD72DFCwsPp0qUL3377LT4+PmpHE0IIIVKFVFgiXQQHBzN06FDy5s/L8nXL8ejrQaEthcjeJrsU+plIxL0IlGiF8t7eakdJFdYWFvSqUIGz/foxtlYttqxbh1/+/PTr148HDx6oHU8IIYT4YNKyL9LUy5cvmT17NrPmzCIyJhLXDq64f+6OWTYztaOJ9/B8/3OCRgRxsk8fCubIoXacVBcWEcGPJ0+y4O+/0cfE0KdvX0aMGEEOE7xXIYQQWYMU+yJNhIeH8/333zN56mRevX6Fa2tXPLp4SJ/8TO7Bkgc8+fkxz74Zq3aUNPVSr+eHEyf44eRJDFoto7/+mq+++kqm7BRCCJHpSLEvUpXBYGDFihWM+noUwcHBuDRzwaO7B5Y5ZCEsU3Bz6E04GcGt4SPUjpIunoeHM/3PP/np33/J5eXFrDlzaN68uUwHK4QQItOQztIi1Zw/f55KVSrRrVs3oktGU3hTYXxG+0ihb0J013R4OziqHSPdONva8t3HH3O8d2/8rK1p2bIldWrX5sKFC2pHE0IIIVJEin3xwUJDQxk0aBClS5fm8pPL+P3oR54pebDytlI7mkhFBr2BiAcRFHV3VztKuiuQPTsbP/+cDZ9/zr3Ll/noo4/o3bs3wcHBakcTQgghkiTFvnhviqKwbt06/Ar68f2P3+PRzwO/tX5kK5NN7WgiDeiD9KBAldy51Y6imvoFCnCsVy8m1qvHulWr8Mufn3nz5hEdHa12NCGEECJBUuyL9+Lv70/turX57LPPiCoSRaGNhfDo7IHWQt5Spkp3QwdA3fz5VU6iLktzc/pWqsTpfv1onj8/gwcPpnzZspw5c0btaEIIIUQ8UpmJdxIeHs7XX39N8RLF+efaP+Sbnw/fGb5Y5pR++aZOF6jD3NKMnA4OakfJENzs7JjTpAn7u3Uj6skTypUrx9ChQ3n9+rXa0YQQQggjKfZFiu3atYuChQsybcY03Lq6UWB9ARyrZJ3BmlmdPlCPs6VMPfm2Mrlycah7d76tVYvv58+nWJEiHDhwQO1YQgghBCDFvkiBV69e0aNHDxo3bkyYZxgFNxTEs5cnWmt5+2Ql4QHh+Dq7qB0jQ7IwM2NwtWoc792bXGZm1K1bl549exIaGqp2NCGEEFmcVGsiSX///TclSpVgxZoV+HztQ96FebH2kdbdrCY6NJroZ9GUzJlT7SgZWl5XV7Z37MjsRo34ddUqihYuzO7du9WOJYQQIguTYl8kKDo6mnHjxlGlahWCbYIpsLYAbi3cZDGhLEp/Qw9A9Tx51A2SCWi1Wr4oV47jX35JAVtbGjZsyMCBA9Hr9WpHE0IIkQVJsS/iuX79OpWqVGLCpAnk6JaD/EvzS2t+FqcL1IEGauXLp3aUTMPHyYnf2rdnWsOGLFm0iArlynHlyhW1YwkhhMhipNgXRoqi8NNPP1GiVAmuPLpCgWUF8OzlicZCWvOzOt0NHVaW5thbyUJp70Kj0dCrQgUOdOtG5NOnlCldmsWLF6MoitrRhBBCZBFS7AsAnjx5QtNmTenZsyf2DezxW+OHXXE7tWOJDEJ/TU92a3k/vK9iHh4c7NaNz4sX58svv6RF8+aEhISoHUsIIUQWIMW+YPfu3RQpXoR9f+0j7+y8+Hzjg5mtmdqxRAahKArhgeHkd3VVO0qmZmtpyezGjVndti1H9u+nZPHiHDp0SO1YQgghTJwU+1mYoihMnjyZTz75hBi/GAqsK4BTDSe1Y4kMJuppFIbXBkp7eqodxSQ0LlyYY716kdfGhjp16jB+/HgMBoPasYQQQpgoKfazqFevXtGydUu++eYb3Hu44zvXFws3C7VjiQxIH/hmFpnaMjg31Xg6OLC1QwdG1azJ+PHjaf7pp7x8+VLtWEIIIUyQFPtZUGBgIOUqlGPHHzvIOzPvm0G4WhmEKxKmC9SBFir6+KgdxaSYabUMr1GDdZ99xpH9+6lQrhz+/v5qxxJCCGFipNjPYnbv3k2ZcmW48/oOfiv9cKrlpHYkkcHpbuiwtbTE3EzGcaSFBgUKcLB7dzShoZQvV47t27erHUkIIYQJkWI/i1AUhalTp/LJJ5+gLaYl/8r82OS1UTuWyAT013TktLVXO4ZJy+fqyr4vvqCGtzfNmjVj3Lhx0o9fCCFEqpBiPwt4/fo1bdu1ZdSoUbh/4Y7vbF/Ms5mrHUtkAkqMgi5IT6Hs2dWOYvKyWVmxqnVrvqldmwkTJkg/fiGEEKlCKj4Td/PmTZo0a8K1m9fwne6Lcx1ntSOJTCTifgRKlEJ5b2+1o2QJWq2WodWrU9zDg55btlCpQgV2792Lj4yXEEII8Z6kZd+EnThxgtLlSnPr5S38lvtJoS/emS5QB0A9Pz+Vk2QtDQoUYH+3boQHB1OxfHnOnz+vdiQhhBCZlBT7Jmr37t3UrlsbJbfypn9+fumfL96dPlCP1lwj3XhU4Ofmxt4uXXA3N6da1ars379f7UhCCCEyISn2TdDatWtp3KQxNmVtyLswL+aO0ltLvB9doI5sllZotfKrQg3u2bKxs1MnKuTMScOGDVm9erXakYQQQmQy8i+4iZk/fz7t27fH6WMn8szIg9ZafsTi/emu6ciVzVHtGFmavZUVv7ZrR9vixenYsSNTp05FURS1YwkhhMgkpMnXRCiKwpgxY5g0aRI5OubAa6AXGo0slCXenyHCQMT9CIoVc1c7SpZnYWbGwqZNyeXgwKhRo7hz5w4LFizATNY+EEIIkQwp9k1ATEwMffr04ccff8RroBfunaQ4Ex9Of0sPBqgkM8FkCBqNhlG1auHl4MDgJUt48vgxa3/9FUtLS7WjCSGEyMCk2M/kIiIi+LzD52zZvIXcY3Pj2tRV7UjCROhuyEw8GVGnMmXIbm9P540bad2qFRs2bsTKykrtWEIIITIo6dCdiYWFhfHxJx+zbfs2fGf4SqEvUpU+UI+5pRYvR+mzn9E0LFiQtW3bsmf3blo0b45er1c7khBCiAxKiv1M6sWLF9SoVYNj/xwj78K8ONV0UjuSMDG66zqcLGXK1oyqrp8f69u149CBAzRr2hSdTqd2JCGEEBmQFPuZ0KtXr2jQsAGXAy+Tb0k+spXJpnYkYYJ013TkcZKF2DKymvnyseGzz/jryBEaf/IJr1+/VjuSEEKIDEaK/UxGp9PRuGljzl48i+8CX2wL2aodSZigmLAYooKjKOnhoXYUkYxqvr5sat+ef06c4JOGDQkLC1M7khBCiAxEiv1MJCIiguYtmnPsxDF85/liV9RO7UjCRMUOzq2aJ4+6QUSKVM6dm9/at+fsv//ycf36hIaGqh1JCCFEBiHFfiYRHR1Nu8/bse/APvLMyoP9R/ZqRxImLLbYr5M/v8pJREpV8PFha4cOXD5/nk+bNZNBu0IIIQAp9jOFmJgYOnfpzPbt28kzPQ8OFR3UjiRMnP6GHksrcxysrdWOIt5BmVy5WNeuHSeOHeOzdu2Ijo5WO5IQQgiVSbGfwSmKQu/evfn111/JPTE3TtWd1I4ksgDdNR3ZrWQ8SGZUKXduVrZuzY4dO+jRoweKoqgdSQghhIqk2M/AFEVh8ODBLF26FJ8xPjjXl5lRRNpTFAXddR35XGXdhsyqQYECLPr0U1asWMHw4cOl4BdCiCxMVtDNwL755hvmzZuH9whvXJtI4SXSR3RwNDGvYihd0lPtKOIDtClRgmfh4YycORM3NzdGjBihdiQhhBAqkGI/g5o7dy5TpkzBa6AX2dtkVzuOyEJiB+fWzJtX5STiQ/WuWJFnOh0jR47ExcWFHj16qB1JCCFEOpNiPwPatm0bX331FTk65cC9k7vacUQWowvUgRaq5M6tdhSRCkbVrMmz8HB69+6Nq6srLVq0UDuSEEKIdCR99jOYU6dO0e7zdjjVdsKrv5facUQWpL+hx9bSAgtzaQswBRqNhukNG9KsSBE6tG/P6dOn1Y4khBAiHUmxn4HcuXOHTxp/gkU+C3JPyI1Gq1E7ksiCdNd0eNhmUzuGSEVarZYfmjWjsJsbTRs35sGDB2pHEkIIkU6k2M8gQkJCqFajGq/MXpFnVh601vKjEelPMSjobuoo6OamdhSRymwsLFjbti3odDRr0oTw8HC1IwkhhEgHUlFmAAaDgU4dO3Dn9h00jhqQBn2hksj7kSiRCuVy5VI7ikgDHtmysbZtWy5fusQXXbvKlJxCCJEFSLGfAYwZM4Y/du+m72egv67Dv70/uus6tWOJLEgX+OZ9V9fPT+UkIq2U8vRk8aefsn7DBiZOnKh2HCGEEGlMin2VrV+/nsmTJzNtCCz8Fo6vVjALiyagcwAvj75UO57IYnQ3dGjMNBRzl1mgTFmzIkX4ulYtxo4dy8aNG9WOI4QQIg1Jsa+iM2fO0LVrZzo00TD0izfbKpSEoD0KORwN3Bh8g8drHstX7SLd6AJ1ZLOyQquVXw2mbmj16rQqXpzOnTpx6tQpteMIIYRII/IvukoeP35Ms2aNKJY/mh8nKGj+00/fIzvcPQAVisP92fe5M+kOhiiDemFFlqEP0JPL3kHtGCIdaDQaFjRtSmE3N1q1aMHz58/VjiSEECINSLGvgpiYGD77rA3REcFsWRCDjXX8fczN4e/10K0lhGwLIbBPINEvotM/rMgyDJEG9Pf0FJEuPFmGjYUFK1q14mVICF27dJFvEYUQwgRJsa+CyZMnc+TIUdbNjMYrmbpq6SRY8DW8Pv8K/47+6IP06RNSZDn623owQCUfH7WjiHTk4+TED02bsm37dubOnat2HCGEEKlMiv10duTIEcaPH8fYPgo1yqfsmH7t4eDPoHkWhX8nf0JPhqZtSJEl6QPffJCsmz+/yklEevukUCH6VarE8OHDOXnypNpxhBBCpCIp9tPR06dP+fzzNtQop+Hr3u92bM3ycO13BRdrA4H9Anm68WnahBRZli5Qh5mFltzOzmpHESoYW7cuH+XMSZtWrXj27JnacYQQQqQSKfbTicFgoHOnjkRFhLB6ugEzs3c/h0/ONwN3SxWAu1Pvcnf6XZRo6WMrUocuUIeTZQIDSESWYGFmxs8tWxL2/DldOneW/vtCCGEipNhPJ3PmzOGP3XtY9V0Mnjne/zzW1nB2C3zeCJ5ueMqNATeICYtJvaAiy9IF6MjjJK36WZm3kxOLmjZlx86dzJ49W+04QgghUoEU++ng/PnzjBo1kiFd4eNqqXPONTNh2lcQdiqUgE7+RNyLSJ0Tiywp5lUMUU+jKO7hoXYUobKPCxZkQOXKjBw5kn///VftOEIIIT6QFPtpTKfT0b59WwrnVZg8KHXPPbw77FoEMY8j8e/gz6uzr1L3AiLL0N3UAVA1Tx51g4gM4ds6dSjm7k6nDh3Q6XRqxxFCCPEBpNhPYyNHjiQw8DprZ8RgZZn6529YDa5sU8imNXCt1zVCtoek/kWEyZOZeMR/WZiZseTTTwkKCmL06NFqxxFCCPEBpNhPQ3v27GH+/PlMH2KgqF/aXSd/brh3UKGgD9wef5v78++jGGRwnUg53Q0dFlZmONnYqB1FZBAFs2dnbO3azJ07l4MHD6odRwghxHuSYj+NhISE0LVrR+pX0dKvfdpfz84WruyEZrXh8crH3Bxyk5hwGbgrUkZ3TYebpa3aMUQG07tCBarlzUvXzp0JCwtTO44QQoj3IMV+Ghk0aCD68Gcsn2JAm06vslYLW7+HsX0g9K+XXOtyjciHkelzcZGp6QJ15HN1VTuGyGC0Wi0LmzQh5OlThg0bpnYcIYQQ70GK/TSwd+9eVq9ew6zhHzbN5vsa1x9+mwtRd/X4d/Dn9cXX6R9CZBpRIVHEhMbwkaen2lFEBpTb2ZkJdeuyZMkSDhw4oHYcIYQQ70iK/VQWHh5O717dqVVBS5fm6uVoXg/O/aZgExPDte7XeLZHVsQUCdPdeDPbSk1fX5WTiIyqa5kyVMubl25du0p3HiGEyGSk2E9l48eP58HD+ywZZ0CjUTdL0fxwZ79Cbg+FW6Nv8WDxA1kVU8SjD9SDFqpJsS8SodVqWdCkCU8fP2bMmDFqxxFCCPEOpNhPRefOnWPWrJmM7WPAL4/aad5wcoAbe6B+FXj00yOCRgZh0BvUjiUyEF2gDhtLCyzNzdWOIjKwPM7OjKhenQULFnDhwgW14wghhEghKfZTSUxMDK1aNsfeVmFQJ7XTxKXVwp6lMLQrvDj4gmvdrhH1NErtWCKD0F3T4W5jr3YMkQl8WbEi+V1d6dO7t3xLKIQQmYQU+6nkxx9/5MbNW7wMU8hZw4wt+9ROFN+M4bB6KkTc0OHfwZ9w/3C1IwmVKQYF3U0dBd3c1I4iMgFLc3NmNGzIsRMnWLVqldpxhBBCpIAU+6ngyZMnjBo9mjqtPmPcio3YOHnTcqCGim3hSQZb0LZ9E/hnnYKFLpqArgG8OPRC7UhCRZEPI1EiFMrmyqV2FJFJVPf1pVXx4gwbMoTnz5+rHUcIIUQypNhPBSNGjMCgQIchoylesQpzdx6i3YBhnLpigXdtc8bMVzthXKWLwO29Cp4uCjeH3uTRikfylXwWpQt8MxNP3fz5VU4iMpNJ9eujf/2ar7/+Wu0oQgghkiHF/gc6duwYK1as4PPBo3BwfrMokYWlFa2+HMS8XYcpVLYSExeBV00NJ8+rHPY/3Fzg1j6oWgYeLHjA7bG3MUTKwN2sRn9Dj8ZMQ8mcOdWOIjIRj2zZGFWjBosXL+b06dNqxxFCCJEEjSJNuu/NYDBQpmxZXkUZmLJ+J2ZmZvH2URSF47t3sGzi14S9eE6jGgbWzwIbGxUCJ6LvBPhhHdgVsyPvnLxYOFuoHUmkk6DRQUT/Gc7dEaPUjiIymeiYGGr89BN2uXJx4uRJtOm1VLgQQoh3Ir+dP8D69es5d/YsXUaNS7DQB9BoNFRp2JQFe47x8edd2HlYg1sVM37akM5hk/D9GFgyFnRXXxPQPsC4yJIwfboAHV72DmrHEJmQuZkZMxs25J9Tp1izZo3acYQQQiRCWvbfU2RkJAULFSK7rx8jf1iR4uMCL55n8ZihBF29TOG8sHMR5PVJu5zv4tgZqNtdQ6SiwXe6L45VHNWOJNKQIcrAucrnaFG4KD+3bq12HJFJdVi/ngthYQRcv461tbXacYQQQrxFWvbf05IlS7hz+zafD3637g/5i5dk2qbddP92Mjcf2lKwkRm9x4EhA3SXr1IabuxWcLM3cGPgDZ6sfSIDd01YxO0IMEBFnwzyaVNkSmNq1+begwcsWrRI7ShCCCESIMX+ewgNDWX8hAnUbN4GH7+C73y8mZkZDdt35fs9x6hQrxFL1oNbZS2/H0mDsO/IMwfcPQBli8K9Wfe4O+UuSpQU/KYotrtWgwIFVE4iMrMC2bPTsVQpJk+cyMuXL9WOI4QQ4i1S7L+HWbNmERb2irb9hnzQeZxzuPPVnMV8u3Qt5vaeNP5SQ7X28OxF6uR8X5aW8O9G6PwpBG8JJrBfINGh0eqGEqlOF6jDzEJLbmdntaOITG5EzZqEv37NjBkz1I4ihBDiLVLsv6PHjx8zc9YsGnboiltOr1Q5Z6mqNZm36wgtvxzI8fNm5KxhzuTFqXLqD7LiO5g7El6dDSOgYwD6O3q1I4lUpA/U42gpfazFh/N0cODL8uWZPWsWDx8+VDuOEEKI/5Bi/x1NnDgRrZk5zXv0S9XzWlnb8NmA4czZcRC/kuX4Zh741NZw5kqqXuadDewE+34C5WkkAR38CfsnTN1AItXoAnTkdnRSO4YwEQOqVMHazIzx48erHUUIIcR/SLH/DgIDA1myZAnNe/Yjm1PadH3IldeP8as2MXDGQl7onCjXRkurgRAZmSaXS5E6leDaLgVHSwPX+14neHOwemFEqogJjyHycSTFPTzUjiJMhJONDUOqVGHp0qVcu3ZN7ThCCCH+R4r9d/D1N9/g5Jadhh2+SNPraDQaqjdpwcI9x6jbqj2/7dPgWtmMVdvS9LJJyu0F9w9CsXxwZ/Id7s26hxIjA3czK/2NN12yqubJo24QYVK6lyuHp4MDY8eMUTuKEEKI/5FiP4X8/f3ZuGEDrfp8hZV1+ix/a+/oRK/x0/ju1+24eOan80go0QzuqNQl1toaLm6H1h/Dk1+fcGPgDWJexagTRnyQ2Jl46ubPr3ISYUqsLSwYWKkSGzZuJDAwUO04QgghkGI/xaZPn45LDndqftoq3a9doFQZZm7ZR5eR4wi4bU2+BmYMnKze3Pwb5sCUQRD2TygBnQOIuB+hThDx3vQ39FhYmuFia6t2FGFi2n/0EW729kybNk3tKEIIIZBiP0Xu3bvH6tWr+aRTdywsrVTJYGZuTpMuPVmw+yila9Rn/mrIUVXLgROqxGFUT9i+EGIeRuDf3p9X516pE0S8F901Ha5WUuiL1GdjYUHf8uVZtXIl9+/fVzuOEEJkeVLsp8DcuXOxtLahftuOakfBLacXIxYuY9SilWDpTr3uWmp3gVAVau3GNeHyVgU7TQzXel4jZFdI+ocQ70V3XUdeFxe1YwgT1bVsWWwtLJg9e7baUYQQIsuTYj8Zz58/Z/GSJdRv1wlb+2xqxzEqW6seC3Yf5dNufTjyrxk5qpoz6+f0z+GX583AXb9ccHvMbe4vvI9ikIG7GVnU8yiiX0ZT2tNT7SjCRDlYW9O9bFmWLF5MSIg0AgghhJqk2E/GokWLiIqKolGn7mpHicfKxpYOQ0Yza9t+8hQpxdAZkLeehkvpPOudvR34/w6NasDj5Y+5OewmMToZuJtR6QPfzMRT3ddX5STClPWuUAFDdDQLFixQO4oQQmRpUuwnQafTMXfePGp+2gbn7DnUjpMoH7+CTFqzlb6TZ/PkZTZKtTDj86EQHZ1+GbRa2LkYRveCl3++5FrXa0Q+VnFxAJEoXaAONFLsi7TlZmdH548+Yv7cuYSFyWJ8QgihFin2k7By5UpCgoNp+kVvtaMkS6vVUrtlOxbuPUaNZq34dRe4VDJj/e/pm2PyIFg/EyJv6fFv78/ry6/TN4BIlu6GDmsrC6wtLNSOIkxcv8qVefXqFT/++KPaUYQQIsvSKIoiHawTEB0dTYGCBclZoChD5i5RO847u3LqJIvHDOP+zUBKF4Edi8AzHb+cuOAPVTpqeB0BeSbmwble2qw4LN5dQKcA3O5bcH7QILWjiCyg5+bN/PvyJYE3b6LVSvuSEEKkN/nNm4jNmzcTdPMmn3bvq3aU91KkbAVmbd1PhyGjuRhoSZ665oyYkX5z85coBHcPKPjkUAgaGcTDnx4inyvVpygKuhs6Cri5qR1FZBE9ypUj6PZtdu/erXYUIYTIkqTYT8T3P/xA0XIVyVeshNpR3puFpSXNe/Rj/u9HKV65BtN/Bs8aWo6eSp/rOznAzb1QuyI8XPyQW6NvYdCrtBKYACDyUSQGvYEyXl5qRxFZRNlcuSjp5cUP33+vdhQhhMiSpNhPwLVr1/jzyBHqtemgdpRUkSOXN6MXr2L4wmVEatyo2VnLxz3gVTp0p9dq4cByGNwZnu9/zrUe14gKjkr7C4sE6a+/mYmnjp+fyklEVqHRaOhWpgy///EHQUFBascRQogsR4r9BCxduhQHJ2cq1P9E7SipRqPRUKFuQxbsPkbjzj3Ye1xLjqrmLFyTPtefPRJWTgH9dR3+HfwJvxaePhcWcehu6NCYaSgjc+yLdNSqWDEcrK1ZsiTzjX8SQojMTgboviUyMhKvXLmo0LAZX4yeoHacNHPL/zKLvh1G4MVz+OV+M4C3YDrMxPjvRajRRYM+Bny/88WphlPaX1QYBX0dRNSRcO6NGKV2FJHFjNq9mw0BAdx78ABra2u142QoiqIQExNDdHrOlyyEyLTMzc0xMzNDo9GkaH8p9t+yceNG2rRpw5wdh/DxK6h2nDRlMBjYv3Etq6ZPIDJCR8cmMfw0AczN0/a6T0Lgoxbw4Al4DfQiR8ccKX7Dig9zpfUVcofac7JfP7WjiCwmMDiYsgsXsmrVKjp27Kh2nAxBURRevHjB06dPiYmRhQiFEClnZmZGjhw5cHR0TLaGkmL/LfXq1efesxdMXrtN7Sjp5kXwU1ZOn8Cf23/D0cGM5ZNiaF4vba8ZHQ01OsHxs+DaxBXvr73RWkivsrSkRCmcq3qOZgUKs6JNG7XjiCyo+erV6FxcOHHypNpRMoSHDx/y4sULHBwccHBwwNzcXBo+hBBJUhSF6OhoQkNDCQ0NxcnJiZw5cyZ5jBT7/xEUFETevHnpO2UOtVu0VTtOurv4918sHjOcx3dvU764wvYfIIdr2l6z91hYshHsS9qTd2ZezJ3T+GuFLEx3U8fV1lf57uOP+bJiRbXjiCxox9WrdFy/nvPnz1OiROad6Sw1xMTEcP36ddzc3HCTqXCFEO8hODiY4OBg/Pz8MDMzS3Q/aUr9j6VLl2KXzYEqDZuoHUUVxStWZe7OQ7QbMIxTVyzwrm3OmPlpe83F4+GHbyH80iv8O/qjC9Kl7QWzMH3gm5l46stMPEIlDfz8cLGzY82adJoZIAOLiopCURTs7OzUjiKEyKTs7OxQFIWoqKRnOZRi/3+io6P5eflyqjVpjpWNrdpxVGNhaUWrLwcxb9dhCpWtxMRF4FVTw8nzaXfNLz+DIytA8zyKgE4BhJ4ITbuLZWG6QB1aCy35XNP46xohEmFpbk7zwoVZu3o1hvRa4S+Dk247Qoj3ldLfH1Ls/8+uXbt49PChycyt/6Fy5vZlzLJ1fDVnMa+iXKj8uZZmfUCvT5vrVS0DgX8ouNoYCBwQyJP1T9LmQlmY7oYORwsrtWOILK51iRLce/CAP//8U+0oQgiRJUix/z+//voreQsXI0+hompHyTA0Gg1VGjZl4Z5jfPx5F3Yc1uBaxYylG9Pmerk84N5B+Kgg3Jt+jztT76BEyZCS1KLz1+Hj6KR2DJHFVfD2JreLi3TlEUKIdCLFPqDT6dixcycVGzRWO0qGZJfNgW7fTGLqht9xz12IHmOgSGMIupv617K0hDOboX0TCN4UTGD/QKLDZO7pDxWjiyHycSTF3N3VjiKyOI1GQ+uiRdm4YQP6tPqqUAghhJEU+8Du3bsJf/2aSg0aqR0lQ8tfvCTTNu2m+7eTufnAlgKfmPHlOEiLrrerp8PMYfDqTBgBHQPQ35Wi4EPob+pBgWp58qgdRQhalyjBy9BQfv/9d7WjCBPzxRdfoNFocHV1JSIiQu04STp37hy9e/emSJEiODg4YGlpiYeHB/Xq1WPWrFk8ffpU7YjCREixD2zatIk8hYrg6ZtP7SgZnpmZGQ3bd2Xh7r8oX+8TFq8Ht8pafj+S+tca0hX2LAHDk0gCOgQQdios9S+SRehvvPmwVCd/fpWTCAEFs2enpJcXq3/5Re0owoSEhYWxYcMGNBoNz549Y+vWrWpHSpDBYGDo0KF89NFHLF26FA8PD7744guGDRtGkyZNePjwIUOHDsXX15f79++rHVeYgCxf7Ov1erbv2EHF+tKq/y5c3D0YMmcJ3/y0FnN7Txp/qaFae3j2InWvU68KBOxQcDCP4fqX1wneGpy6F8gidIE6zC3NyG5vr3YUIQBoU6wYu3bt4vnz52pHESZi/fr1vH79msGDB6PValm2bJnakRL09ddfM2vWLEqXLo2/vz8HDx5k7ty5TJ48mZ9++olLly5x+vRpKlWqhE4n01GLD5fli/29e/fyKiyMSh9Lf/338VG1mszbdYSWXw7k+HkzctYwZ/Li1L2GrzfcOwBFfOHOxDvcm3MPJUYG7r4L3XUdrpY2ascQwqhF0aJERkWxc+dOtaMIE7Fs2TLMzc0ZPnw4tWrV4sCBA9y+fTvR/Y8cOUL16tWxs7PD1dWVtm3bcvfuXWrWrJnglIaKovDzzz9TpUoVHBwcsLW1pWzZsvz8888pznjt2jVmzJhB9uzZ2b17N/kT+ba1dOnS7Nu3jzz/6Xp569YtNBoNXbp04erVqzRv3hxXV1c0Gg23bt0C3kwjPnv2bEqWLImNjQ2Ojo7UqlWLHTt2xLvGuHHj0Gg0HD58ON5zK1asQKPRsGLFigSvf/nyZRo1aoSTkxP29vbUr1+f06dPxzvPw4cPGThwIH5+ftjY2ODk5EThwoXp3bs3L1++TPHrJj5Mli/2N23aRG6/guTKKwsNvS8raxs+GzCcOTsO4leyHN/MA5/aGs5cSb1r2NrCxe3Qoh48Wf2EG4NvEPM6JvUuYOJ013X4urioHUMIo5wODpTOlYvt27apHUWYgCtXrvD3339Tv3593N3d6dSpEwaDgeXLlye4/969e6lbty7//PMPrVq1omfPnty+fZuqVavy4sWLePsrikL79u3p1q0bT58+5fPPP6d79+68fv2abt26MXTo0BTlXLlyJTExMfTq1Yvs2bMnu7+5efxV5QMDA6lYsSJPnz6lS5cudO7cGUtLSxRFoVWrVgwZMgS9Xk/fvn35/PPPOX/+PE2bNmXOnDkpypicmzdvUqVKFXQ6HV9++SVNmzbl0KFDVK9enZMnTxr3Cw8Pp0qVKixYsIB8+fLRv39/unTpQoECBfjll19kTEI6iv8uykIiIiLYum0bDTt2VzuKSciV14/xqzZxdOcWlk3+lnJtQmlRJ4Y1M97MsvOhtFr4bT5M+AHGfR9KQJcA8s3Ph1VOmTs+KdEvool+Hs1HBXKqHUWIOD7282PBnj1ERkZimRq/JESWFdtlp2PHjgC0aNGCPn36sHz5csaMGYNW+/9tmzExMfTs2ZOYmBgOHTpE1apVjc917tyZVatWxTv/0qVL+fXXX+natStLlizBwsICgMjISFq1asWsWbP47LPPKFOmTJI5T5w4AUCtWrXe+16PHTvGmDFjGD9+fJztq1atYtu2bdSoUYO9e/ca/06NGjWKMmXKMHz4cJo1a0bevHnf+9oAR48eZeTIkXz33XfGbZ07d+bjjz+mR48eXLhwAYADBw4QFBTEoEGD4n3QePXqlfE1FGkvS7fs79+/n7DQUCrJlJupRqPRUL1JC77fc4y6rT5n0z4NLpXNWJWKjXdj+sCW+RB9L4KA9v68uvAq9U5ugnQ33vT5rO7rq3ISIeJqWLAgYa9eceRIGozwF1lGVFQUv/zyCw4ODnz66acA2Nvb07x5c+7cucP+/fvj7P/XX39x+/ZtmjRpEqfQB5g0aRJmZmbxrrFw4ULs7Oz4/vvv4xSplpaWTJ48GXizXk9yHj16BICnp2e85w4fPsy4cePiPBLqYuPh4cHXX38db/vKlSsBmD59epwPzz4+PgwePJjo6OhUWd/Cyckp3vUbNGhAnTp1uHjxYrzuPDY28buQ2tvbY2UlDXXpJUu37G/atAnvfH54+xVUO4rJsXd0otf4adRq3oZF3w6l88gAZv4MOxeDTyo0MDerAxc2K1T8zMD1HtfJPTY3Lp9IN5WE6AJ1oIEaH9iaI0RqK+buTi5nZ3bs2EG9evXUjiMyqW3btvH06VO6deuGtbW1cXunTp1YvXo1y5Yto379+sbt58+fB4hX6AN4e3vj4+NDUFCQcVt4eDgXL17E09OTadOmxTsmKioKAH9//w+6j8OHD8drrQeoWbNmnD+XLFkywW/Czp49i62tLeXLl4/3XOw3CefOnfugjAAfffQR9glM9lCtWjUOHDjA2bNnKVOmDNWrVydnzpxMnTqV8+fP07hxY2rUqEHhwoUTHBMh0k6Wbdk3GAzs2LmTcnU+ljddGipQqgwzt+yjy8hxBNy2Jl8DMwZOTp25+Qvng7sHFfLmVLj17S0e/PAAxSADd9+mv6HH2socW+kmITIYjUbDx/nzs2PbNhRF/u6K9xPbhadTp05xttepUwcvLy+2bdvGs2fPjNtDQ0MByJEjR4Lnc39r8cHnz5+jKAr3799n/Pjx8R5TpkwB4PXr18lmjT33gwcP4j03btw4FEVBUZQkvyV4O99/7yuxe8qZM6dxnw+V2PVjt8cOvHV0dOTvv/+mU6dO/P333/Tp04eiRYuSO3dufvjhhw/OIVIuyxb758+fJyQ4mFJVa6odxeSZmZvTpEtPFuw+Suka9Zm/Gtyrajlw4sPP7WAP13bDx1Xh0bJHBI0IwqBLg1W+MjHdNR05rGXKTZExfVygALfu3OHSpUtqRxGZ0N27d9m7dy8ANWrUQKPRGB9mZmbcv3+fiIgIVq9ebTzGwcEBgCdPniR4zsePH8f5c+z+ZcqUMRbjCT0OHTqUbN7KlSsDpGjfxCTWQOng4JDoPcV2H4q9F8A4jiE6Ov4q9UnNlPP26/P2dkdHR+M2Hx8fVqxYwdOnTzl79izTpk3DYDDQt2/fFHV7Eqkjyxb7+/fvx9rGlgKlSqsdJctwy+nFiIXLGLVoJYqlO/W6a6nTBUI/sMu9Vgt//AQju8OLwy+41i2AyCeRqZI5s1MUBV2gDj9XV7WjCJGgar6+2FtbJzg1oBDJWbFiBQaDgapVq9KtW7d4j86dOwPEmXO/ZMmSwJuBrm+7d+8ed+7cibMtW7ZsFC5cmKtXryY4U8+76Ny5M1qtlh9//JHg4NRdN+ajjz4iPDycf/75J95zsX3/S5UqZdzm7OwMkODCXWfPnk30OmfPnuXVq/j/cB89etSY421arZZSpUoxfPhwY5G/ffv2xG9GpKosW+zv27efwmUrYGEpA0TSW9la9Zj/x1E+7daHw/+akaOqObNSPk1xor4bAr9Oh4ibevzb+xN+NfzDT5rJRT2OwqAzUNrLS+0oQiTIytycWnnzskOm4BTvSFEUli9fjkajYeXKlSxdujTeY8WKFVSqVIkLFy5w6tQp4E1ffR8fH3bs2GGcHSfWt99+S0xM/GmdBwwYQHh4OD169Eiwu05QUJBxrvukFChQgOHDh/PkyRMaNmxIYGBggvu9z4eK2A82o0aNMo4jgDfffsyePRtzc3Pat29v3F6uXDngzSw+hv/0rT1x4kSSA3lfvHhhHJQca8+ePRw4cIBixYoZZyS6fPlygt8CxG777/gKkbay5ABdvV7PX38dpU3/YWpHybKsbW3pMGQ01Zu2ZPGYYQydcYrvf9Ww/XuFYgXe/7ztGkGhvApVO8YQ0DWAPJPz4FzHOfWCZzK6wDcz8dRJZOEWITKCevnzM2jnTl6+fBmnC4AQSTl48CBBQUHUqFEjyekku3btyokTJ1i2bBlly5bFzMyMxYsX07RpU2rXrk3btm3JmTMnR44c4f79+5QsWdI4fWSsXr168ffff7Ny5UqOHTtG3bp18fT05PHjx/j7+3Py5EnWrl0bZxGsxEyePJnIyEhmz55NoUKFqF69OiVLlsTW1pYnT55w4cIF/vnnH+zt7eO0xCenY8eObN68mW3btlGiRAkaN27M69evWb9+Pc+ePWPWrFlxXqeKFStSpUoVDh48SKVKlahevTq3b99m27ZtNGnShC1btiR4nWrVqrFo0SJOnjxJxYoVuXXrFhs3bsTGxoalS5ca99u3bx/Dhg2jSpUqFChQAFdXV27evMn27duxtramb9++Kb438WGyZMv+iRMn0Ol0lKhSXe0oWZ6PX0EmrdlK38mzefIyG6VamNF+GCTQhTDFShWGO/sVcrkpBA0P4uGyh1l28J/uhg6NFsrnyqV2FCESVTV3bgwGA3/99ZfaUUQmEts1p0uXLknu17ZtW2xsbPj111/R6d40gDRs2JC9e/dStmxZNmzYwI8//kiuXLn466+/iImJidO3HTCuJrt+/XqKFi3Kzp07mT17Nvv27cPa2pqZM2dSt27dFOXWarXMmjWLM2fO0K1bNx4+fMjSpUuZMWMGO3bswN7enhkzZnDjxg3jVKIpodFo2LRpEzNnzsTCwoIFCxawevVqihcvzrZt2/jqq6/iHbNt2zY6depEYGAg33//PXfv3mXHjh00bdo00evkzZuXY8eOYWNjw/fff8/27dupWbMmR48epUKFCsb9GjRoQN++fQkNDWXz5s3MmTOHU6dO0bZtW06fPk3ZsmVTfG/iw2iULFgFjR49mkU//sRPR8/FWWhDqCv0eQi/zJjEwc3ryWZvxtIJMbRp+P7nMxigdhc48i+4NHTB51sftFZZ6+d969tbRBx6zf0Ro9SOIkSiFEWh6Lx5fN69OzNmzFA7TrrQ6/UEBQXh6+sr3RkykLCwMNzd3SlevHic1WAF3Lp1C19fXzp37syKFSvUjiNI+e+RrFX5/M++ffspVrGqFPoZjIOzK32nzGHi6i045PCl7RANZVvBg4QnF0iWVguHV0H/9vBszzOu97pO1LOo5A80IboAHZ62MhOPyNg0Gg1VvL058gEzlAjxLl6/fk1YWFicbTExMQwbNgydTvdOLepCZHRZrtp9/vw5p0+fokTlampHEYkoUrYCs7bup8NXo7hw3ZI8dc0ZMfP95+af/w0sHQ+6q6/xb++P7roudQNnUEq0gv62nsKJzLssREZSJXduzpw7lyrzgAuRnOvXr5MzZ05atmzJsGHD6N27NyVKlGDJkiUULVqUAQMGqB1RiFST5Yr9Q4cOoSgKJSpJsZ+RWVha0rxHP+b/fpTilWswfRl41tDy1+nkj01It1ZwfDWYhUUT0DmAl0cTn0PYVETcjUCJVijv7a12FCGSVTVPHmJiYhKcDlGI1Obl5UXr1q05e/YsP/zwA8uXLycyMpKhQ4fy119/YWdnp3ZEIVJNlpuN58CBA3jm9iW7pwxYzAxy5PJm9OJV/HNgNz+NH0WNTsHUq2xg01ywf8ffxRVKQtAehY9aKtwYdAOvwV7kaJ/DZFdQ1t148w1GPT8/lZMIkbx8rq64Ozhw5MgRGjb8gME6QqRA9uzZWb58udoxMpU8efJk2ckuMrss17L/998nKVi6nNoxxDvQaDRUqNuQBbuP0ahTD/Ye15KjqjkLE58GOFEe2eHuQShfHO7Puc+diXcwRJnmiru6QB1acw0Fs2dXO4oQyZJ++0IIkTayVLEfERHBxYsXyFespNpRxHuwsbOjy8ixzPhtN575i9J/EhT4GAKC3u085uZwcgN0awkh20MI7BNI9IsPmOszg9IH6nGwlFk+ROZRNU8e/j19OsHVOYUQQryfLFXsX7x4kaioKCn2MznfwsWYumEXvcZN416wPUWbmvHF6Hefm3/pJFjwNby+8Ar/jv7og/RpE1gluoBwfBxkgSKReVT08SEmJsa40qkQQogPl6WK/X///Rczc3PyFCqidhTxgbRaLfXbdWThnmNU+eRTlm8BtypmbDvwbufp1x4OLgOeReHfyZ/Qk6YxE4hBbyDiYSRF3d3VjiJEihXMnh1bS0sp9oUQIhVlqWL/1KlT5ClQGEsr6dpgKpzcsjNw+gLGrdiAjZM3zftrqNgWnoSk/Bw1y8P13xWcrQ0E9gvk6canaRc4neiD9KC8WZlUiMzCTKulpKenFPtCCJGKslSx/8+/p/AtWkLtGCINFK9Ylbk7D9FuwDBOXbHAu7Y5Y+an/HifnHDvAJQsAHen3uXu9Lso0Zl31gFd4JuZeOrKTDwikynl4cGpf/5RO4YQQpiMLFPsh4eHc/XKZemvb8IsLK1o9eUg5u06TKGylZi4CLxqajh5PmXHW1vDuS3wWSN4uuEpNwbeICYsJm1DpxHdDR3mlma4Z8umdhQh3slHnp7cCAri+fPnakcRQgiTkGWK/XPnzhETE0P+YtKyb+py5vZlzLJ1fDVnMa+iXKj8uZZmfUCfwvG3a2fCtK8g7N9QAjr5E3EvIm0DpwH9dT0uljZqxxDinZXw8ADg/PkUfkoXwsQZDAaKFSvG5MmT1Y6Sptq1a0ebNm3UjmGSssyiWqdOncLSygpvv0JqRxHpQKPRUKVhU0pVrcm6edPZsWY5rlW0zBsZQ/fWyR8/vDsULwjN+kfi38GffHPyYf+RfdoHTyXh18Ip7JxT7RhCvLP8rq5YW1hw7tw5atasqXYc1dy5c4fg4GBVM7i5ueHj46Nqhoxq0aJFHDx4kJMnT3L37l06d+7MihUrEtx33759jB8/njNnzmBlZUWdOnWYOXMmefLkSdG1fv31V+7evUu/fv2M2/79919WrlzJoUOHuHXrFq6urlSsWJFJkyZRoECBZM/5559/MnPmTM6ePcvTp09xcnKiVKlSfPvtt1SpUiXOvnv37mX9+vWcPHmSq1ev4u3tza1bt+Kd88WLF/Tt25fff/8dZ2dnvv76a7p16xZnn1OnTlG9enUuX76Mr69vnOdGjBhB2bJlOX/+PCVLSi+M1JSliv08BYtgYWmpdhSRjuyyOdDtm0nUaNaaxWOG0mPMZWavhF2LwNc76WMbVoPLWxXKtTNwrdc1cn+TG9emrukT/ANEh0YT/Syakvml2BeZj7mZGYXd3Tl37pzaUVRz584dChcuSHi4utMB29pac/VqgBT8CZg2bRphYWGUL1+ehw8fJrrfzp07adasGaVLl2bq1KmEhoYyb948qlatytmzZ8megkUPZ8yYQbt27XB0/P+plKdNm8axY8do3bo1JUqU4NGjRyxcuJDSpUvz999/U6xYsSTPee3aNbRaLb1798bDw4Pnz5+zevVqqlevzq5du/j444+N+65du5b169dTunRpPD09Ez3n0KFDOXz4MOPHjycwMJAePXpQuHBhKleuDICiKAwYMIBBgwbFK/QBPvroI8qWLcusWbNYtWpVsq+LSDmNkkXWPi5WvAReRUvRa/w0taMIlcTExLDn15WsnjWZmKhIureM4fsxoE2mM9ur11CmNVwLghydcuDVzwuNmSZ9Qr+HV2dfca37NX5p04YmRWSaWZH5DNi+nQvR0Zw14a48er2eoKAgfH19sbaOO0PcmTNnKFOmDKunQ+G86uS7ehM6DIfTp09TunRpdUJkYLdv38bHxweNRoO9vT2tWrVKsGW/aNGiREZGcvnyZSz/19h4/vx5SpcuzaBBg5g1a1aS1zl79iylS5dm//791KlTx7j9+PHjlC1b1nhOgOvXr1O8eHFatWrF6tWr3/mewsPDyZs3L6VKlWL37t3G7Q8ePCB79uxYWFjQuHFjLl26lGDLvoeHB9OnT6dTp04A1KxZk0qVKvHdd98BsHr1akaMGEFAQAD29gl/Uz5r1izGjh3Lo0ePEt1H/L+kfo/8V5bos28wGLgReB1P33xqRxEqMjMz45MOX7Bw9zHK1/uExevBrbKWP44mfZy9HVzdCc1qw5NVT7g55CYx4Rl34K7uug40UCufvN9F5lTU3Z0rV68SE5Nx/56lh8J5oXRRdR4f+iHj9u3b9OnTh4IFC2JjY4OrqyutW7dOsEi8cOECNWrUwMbGhly5cjFp0iSWL1+ORqOJt/8ff/xBtWrVsLOzI1u2bDRq1IjLly9/WFjg6dOnrFq1irZt23L0aDL/KAC5c+dGo0m60efZs2dcuXKF5s2bxynKS5YsSeHChVm3bl2y19m6dSuWlpZUr149zvbKlSvHOSeAn58fRYsW5erVq8meNyG2trZkz56dFy9exNnu6emJhYVFssfrdDqcnZ2Nf3ZxcSE8PByA169fM3LkSL777rski/h69erx+vVr9u3b9173IBKWJYr9Bw8eoNfryZk7/tdGIutxcfdgyJwlfPPTWsztPWnUS0P1DvDsReLHaLWw9XsY2wdCj7/kWpdrRD6MTLfM70J3Q4eVlTn2VlZqRxHiveRzcSEyKoq7d++qHUW8p3///Zfjx4/Trl075s+fT+/evTlw4AA1a9Y0FoAA9+/fp1atWly+fJlRo0YxePBg1qxZw7x58+Kd85dffqFRo0bY29szbdo0vv32W65cuULVqlUT/BCRFEVROHPmDBMnTqRixYp4eHjQpUsXbt68iYuLy4fePgAREW8md7CxiT9Zgq2tLQ8ePODRo0dJnuP48eMUK1YsRcW2oig8fvwYNze3FGcMDQ0lODgYf39/Ro8ezaVLl+J8g/AuypUrx+zZs7l+/Tp79uxh9+7dlC9fHoApU6bg5eVFx44dkzxHkSJFsLGx4dixY++VQSQsS/TZv379OoAU+yKOj6rVZN6uI2z+cQGblyzAs6aGb3tH83XvxI8Z1x9KFIK2Q/VvBu7OzYddcbv0C50C+ut6clhlrExCvIu8/yu2AgMDUzyIUWQsjRo1olWrVnG2NWnShEqVKvHbb78Zi75p06bx/Plzzpw5Q6lSpQDo2rUrfm+tEfLq1SsGDBhA9+7d+fHHH43bO3fuTMGCBZkyZUqc7QkJCwtj//797Nq1i99//52HDx+SLVs26tWrR8+ePWnUqBHuqbjquLu7O05OTvEK15CQEK5cuQK8+bDj8b8ZqBLi7+9PhQoVUnS9NWvWcP/+fSZMmJDijG3atGHPnj0AWFpa0qtXL7799tsUH/9fc+bM4ZNPPjEOEG7ZsiWfffYZQUFBzJkzh4MHDyb7bYi5uTne3t7G10ekjizRsn/9+nW0Wi05cskgIxGXlbUNnw0YzpwdB8lfohzfzIPctTWcSeL3TIt6cO43BeuYGK51v8azPc/SL3AyFEUh/Ho4+V0z/kBiIRLj4+SEuZmZsaFGZD7/bc2OiooiJCSE/Pnz4+TkxJkzZ4zP7d69m0qVKhkLfXjT/aN9+/Zxzrdv3z5evHjBZ599RnBwsPFhZmZGhQoVOHToUJJ57t+/j5ubGy1atGDr1q20adOGffv2ERwczG+//cYXX3yRqoU+gFarpVevXhw4cIBRo0Zx/fp1Tp8+TZs2bYiMfPPNsE6nS/IcISEhcbrGJMbf35++fftSqVIlOnfunOKMU6dOZe/evSxbtoyKFSsSGRlJdHR0io//r+LFi3P9+nX+/fdfrl+/zqZNm9BqtQwZMoSWLVtSsWJFNm/eTMmSJfH19WXChAkkNGzU2dlZ9ZmoTE2WadnP4ZVLZuIRicqV14/xqzZxdOcWlk3+lnJtQmlRN4Y10yGht03R/HB3v8JHLeHW6Fvog/Tk7JUz2VaLtBb1NApDuIHSXl6q5hDiQ5ibmeHj4kJgYKDaUcR70ul0fPfddyxfvpz79+/HKepevnxp/P/bt29TqVKleMfnz58/zp9jP/jVrl07wes5ODgkmcfGxobKlSvz119/ERISwp49ezA3N8fMzIyqVaumqJvM+5gwYQLBwcFMnz6dqVOnAlC/fn26devG4sWLUzQINbl5VB49ekSjRo1wdHRk06ZNmJmZpTjffz9kdejQgdKlS9OlSxc2bdqU4nP8l7W1NWXLljX++eDBg+zdu5eAgAACAgJo164dS5YsIU+ePHz22Wd4e3vTtWvXOOdQFEX1f0tNTZZo2Q8MDMRDuvCIZGg0Gqo3acH3e45Rt9XnbNqrwaWyGau2Jby/kwPc2AP1KsOjnx4RNDIIg96QvqHfogt800pUWwbnikwun5MT169dUzuGeE/9+/dn8uTJtGnThg0bNrB371727duHq6srBsO7/56MPeaXX35h37598R7btiXyi/p/XFxcOHToEMHBwaxfv57y5cuzatUqateujZubG61bt2bFihU8fvz4ve43MZaWlixdupQHDx7w559/EhAQwJ49e3j58iVarTbeh5q3ubq6Jrma9MuXL2nYsCEvXrxg9+7dSU6NmZKsTZs2ZfPmzcl+45ASMTExDBw4kJEjR+Ll5cWGDRuoXLkyXbt2pVatWvTq1Ys1a9bEO+758+fvNO5AJC9LtOwHXLtO7lLl1I4hMgl7Ryd6jZ9GreZtWPTtUDqPDGDmz7BzMfi8NXW9Vgt7l8Gw6TBz5Quu3btGvrn5sMieNq1EydEH6kELFbyTWURAiAzO19mZPwMC1I4h3tOmTZvo3LlznKkl9Xp9vJlecufOneA3OG9vy/e/BowcOXJQt27d987l6OhImzZtaNOmDQaDgX/++Yddu3axc+dOunbtikaj4eDBg6m+oJu7u7uxm1BMTAyHDx+mQoUKybbsFypUiKCgoASf0+v1NGnShGvXrrF//36KpMJUyzqdDkVRCAsLS3Bg8btYtGgRYWFhDB06FHgzWcp/P4x4enpy//79OMdER0dz9+5dmjZt+kHXFnGZfMu+wWAg6OYNcuZRabJikWkVKFWGmVv20WXkOAJuW5OvgRmDpkBCjVIzhsPqqRBxQ4d/B3/C/cPj75QOdIE67KwsMX+Hr3GFyIjyubpy89atLD/9ZmZlZmYWr/vJggUL4v08GzRowIkTJ+Isovbs2bN4Lb4NGjTAwcGBKVOmEBUVFe96T58+feeMWq2WihUrMnHiRM6ePcu9e/dYsmQJOXLkeOdzvYuZM2fy8OFDhgwZkuy+lSpV4tKlS8aZfWLFxMTQtm1bTpw4wcaNGxPsChXr4cOH+Pv7x3ndnjx5Em+/Fy9e8Ntvv+Ht7f3Br8GzZ88YO3YsM2bMMM7/7u7ujr+/v3Gfq1evxhucfOXKFfR6vXEhLpE6TL5l/969ezLtpnhvZubmNOnSk0oNGrFs8hjm/fIHa3ZqWTfLQJ23fre2bwKF8ylU6xhNQNcAfKf44lTLKV3z6q7p8LLNlq7XFCIt5P3P9JsyI0/m07hxY3755RccHR0pUqQIJ06cYP/+/bi+NXnA8OHDWb16NfXq1aN///7Y2dmxdOlSfHx8ePbsmbHvtoODA4sWLaJjx46ULl2adu3akT17du7cucOuXbuoUqUKCxcuTDRPWFhYgl1GEpItW/K/Q3fs2MH5/y36FhUVxYULF5g0aRIATZs2pUSJEsCbhaR+++03qlevjr29Pfv372fDhg10796dli1bJnudZs2aMXHiRI4cOUL9+vWN24cMGcL27dtp0qQJz549i7eIVocOHYz/P2rUKFauXElQUJDx71LDhg3JlSsXFSpUIEeOHNy5c4fly5fz4MED1q9fH+dcFy5cYPv27cCbb1xevnxpvNeSJUvSpEmTeLm//fZbihcvTuvWrY3bWrZsyYQJE/jyyy/JnTs3S5YsYfbs2XGO27dvH7a2ttSrVy/Z10aknMkX+7FfBeb0yaNuEJGpueX0YsTCZZw6tI8fx42kXvfH1CpnYMtCcPjPt7Cli8DtfQqlWsLNoTfx7OeJexf3dBlspMQo6G/pKZQvd5pfS4i0Fjv95o0bN7JssX/1Zua99rx58zAzM2PNmjXo9XqqVKnC/v37adCgQZz9vL29OXToEAMGDGDKlClkz56dvn37Ymdnx4ABA+KsCvr555/j6enJ1KlTmTFjBhEREXh5eVGtWrV4gzzfFhISwpdffpmi7H/88QfeyXSF/O2331i5cqXxz2fPnuXs2bMA5MqVy1jsFyhQgGfPnjFx4kR0Oh0FCxZk8eLF9OzZM0VZypQpQ4kSJdiwYUOcYj/2m5AdO3awY8eOeMf9t9hPyBdffMG6deuYM2cOL168wNnZmYoVK7J27VqqVasWZ98zZ87Em44z9s+dO3eOV+xfvHiRpUuXcvLkyTjbixcvzvLlyxk3bhxhYWH06dMn3uuwceNGWrRokaIPXCLlNEpyw7wzuRUrVtC1a1fWXQjCwlIWGRIfTh8ezqZFc9m2bBGWlhom9Y9myBdx94mOhlqd4a8z4NLIBZ9vfNBapm2vOf1tPVdaXGFCvXoMqFIlTa8lRFp7FRFBru++Y/Xq1fGmYTQFSS1zf+fOHQoXLkh4uF6ldG/Y2lpz9WoAPj7pP231oEGDWLJkCa9evXqn2WVM0S+//ELfvn25c+cOTk5OasdJM+fOnaN06dJx1lwQSUvq98h/mXzL/sOHD3FwdpFCX6Qaa1tbOgwZTfWmLVk8ZhhDZ5zi+181bP9eodibtUQwN4eja6DvBPhh3TMi7kaQd3ZeLJzTbuCu7sab2RPqJTO7gxCZgb2VFXZWVqk+O0pm4OPjw9WrAarPNe7m5pYuhb5Op4szGDQkJIRffvmFqlWrZvlCH6B9+/ZMmzaN77//nq+//lrtOGlm6tSptGrVSgr9NGDyxf6jR49wdsuudgxhgnz8CjJpzVYOb9nAiqnjKNXiNW0bxrDyuzfFPsD3Y6BkQfhy0msC2geQb0E+bPJ92AwHidEH6tGaayicygvDCKGW7NmyZcliH94U/Gq0qKuhUqVK1KxZk8KFC/P48WOWLVtGaGjoe6/kamq0Wi2XLl1SO0aaW7dundoRTJbJz8bz6NEjHN3SdmS9yLq0Wi21W7Zj4d5j1GjWirU7waWSGRv++P99eraFP1eCNjSKgE4BvDz2MvETfgBdoI5s8g2WMCHudnY8evRI7RgijX3yySf8/vvvDB48mGnTpuHj48Mff/xB9erV1Y4mhEkw+WL/4cNHOEnLvkhjDs6u9J0yh4mrt+CQw5e2QzSUbQUP/ze7WZXScGO3gpu9gRsDb/Bk7ZNkV0V8V7oAHd7ZHFP1nEKoKYetLY+l2Dd5U6ZM4dq1a4SHh/P69WuOHj36QXPpCyHiMvli/9Hjxzi6SrEv0keRshWYtXU/Hb4axfnrluSua86ImW/m5vfMAXcPQNmicG/WPe5OuYsSlToFvyHCQMT9CIpKFx5hQnLY2fHo4UO1YwghRKZm8sV+SEgwDs4uascQWYiFpSXNe/Rjwe9HKV65BtOXgWcNLX+dBktL+HcjdP4UgrcEE9gvkOjQ6A++pj5IDwpUzi3TbgrTkcPePsv22RdCiNRi0sW+wWDgxfPn2Ds5qx1FZEE5cnkzevEqhi9YRqTGjRqdtDTsAeHhsOI7mDMSXp0NI6BjAPo7HzbFnszEI0xRDnt7ngQHyyq6QgjxAUy62H/x4gUGg4FsUuwLlWg0GirUa8iCP/6iUace7Dmuxa2qGQvXwKBOsO8nUIIjCejgT9g/Ye99Hf0NPeaWWjwdpc++MB3Z7ewwGAyEhISoHUUIITItky72Y/+BkGJfqM3G3p4uI8cy47fdeOYrRv9JUKAh5PKAazsVHC0NXO97neDN7zevtu6aDieLtJnSUwi1OPxvkZiwsPf/ICyEEFmdFPtCpCPfwsWYumEXvcZN495Te4o2NWP8D3BnHxTNB3cm3+HuzLsoMe82cFd3XYevs7zPhWmxtXizCN3r169VTiKEEJmXSRf7r169AsDGzl7lJEL8P61WS/12HVm45xhVPvmU5ZvBq7YZkwdC64/h6bqn3Bh4g5hXKeunHB0WTVRwFCVz5kzj5EKkr9hiPzw8XOUkQgiReZl0sR8ZGQmA+f/+wRAiI3Fyy87A6QsYt2IDNk7eNO+v4e5DGNUdwv4JJaBzABH3I5I9j/7Gm8G9VfPkSePEQqQvW0tLQFr2hRDiQ5h0sR8R8aZQsvjfPxhCZETFK1Zl7s5DtBswjH8vWzBrpTmf1oLoBxH4t/fn1blXSR6vC9SBBurKTDzCxEjLvhCZ14YNG3BxcTH2ssjKoqKi8Pb25ocfflDl+iZd7P9/y74U+yJjs7C0otWXg5i36zCFylZiy35wsFEwj4rhWs9rhOxKfDYS/Q09lpbm2FtZpWNiIdJeVu6zf+fOHc6cOaPq486dO++d/9WrV4wdO5aPP/4YFxcXNBoNK1asSHT/hQsXUrhwYaysrPDy8uKrr77Kkj/39NSjRw80Gg2NGzdO8Pnt27dTunRprK2t8fHxYezYsURHp2xdmJiYGMaOHUv//v2xt3/TlTo8PJzvv/+e+vXrkzNnTrJly8ZHH33EokWLEpxeNzAwkFatWuHs7IytrS1Vq1bl0KFDKb6/06dP07hxYzw8PLC3t6dEiRLMnz8/3rXy5MmDRqOJ9+jdu3ec/a5cuUK1atXIli0bZcuW5cSJE/GuOXv2bIoWLRrvdbKwsOCrr75i8uTJ6PUfNtX2+zBP9yumI2OxbyndeETmkDO3L2OWreP47h0snfg1kRHPQFG4PeY2+iA9nn080Wg1cY7RXdeR3cpWpcRCpJ2sWuzfuXOHwoULq/6Nhq2tLVevXsXHx+edjw0ODmbChAn4+PhQsmRJDh8+nOi+I0aMYPr06bRq1YqBAwdy5coVFixYwOXLl9mzZ88H3IFIzKlTp1ixYgXW/5vx6m1//PEHn376KTVr1mTBggVcvHiRSZMm8eTJExYtWpTs+Xfs2EFAQAA9e/Y0brt58yb9+/enTp06fPXVVzg4OLBnzx769OnD33//zcqVK4373r17l0qVKmFmZsawYcOws7Nj+fLl1K9fnwMHDlC9evUkr3/69GkqV66Mn58fI0aMwNbWlj/++IOBAwdy48YN5s2bF2f/UqVKMWTIkDjbChQoYPz/mJgYWrRogYuLCzNmzGD79u00a9aMwMBAHBwcAHjy5AkTJkxgw4YNmJvHL6+7du3KyJEjWbt2LV988UWyr2FqMuliP7Ybj7Tsi8xEo9FQpWFTSlWtybp50/l99c9oNBoeL3+MPkhPnkl5MLMxA0BRFHTXdZTJ/u7/GAuR0Wm1WqwtLFQvetNbcHAw4eHhDJyxkFx51emed+9mIPOG9SM4OPi9iv2cOXPy8OFDPDw8OHXqFOXKlUtwv4cPHzJ79mw6duzIqlWrjNsLFChA//792bFjB02aNHnv+4il1+u5c+dOnAIuq1IUhQEDBtCpUycOHDiQ4D5Dhw6lRIkS7N2711i4Ojg4MGXKFAYOHEihQoWSvMby5cupUqUKXl5exm0eHh5cvHiRokWLGrf16tWLL774guXLl/Ptt9+S/3/dUadOncqLFy+4dOkSBQsWBN58E1GoUCEGDx7M6dOnk7z+kiVLAPjzzz9xcXExXqtGjRqsWLEiXrHv5eVFhw4dEj3f9evXCQgI4Pbt2/j4+NCpUyfc3Nw4ceIEDRo0AGD06NFUr16d+vXrJ3gOJycn6tevz4oVK9K92Df5bjzmFhZoNJrkdxYig7HL5kC3byYxbeMf5ClUBICXf77kWtdrRD5+861VdHA0Ma9iKP2fX6hCmBJbK6ss17IfK1fe/OQtWkKVx4d+yLCyssLDwyPZ/U6cOEF0dDTt2rWLsz32z+vWrfugHOfPn6d///54enqmqL/0tm3baNSoEZ6enlhZWZEvXz4mTpyYYDeT77//nrx582JjY0P58uU5evQoNWvWpGbNmnH2i4iIYOzYseTPnx8rKyu8vb0ZPny4sUHyQ+9vypQpNG/ePMXH/PLLL1y6dInJkycn+PyVK1e4cuUKPXv2jNNC3adPHxRFYdOmTUmeX6/Xs3v3burWrRtnu5ubW5xCP1Zs9qtXrxq3HT16lI8++shY6MObb5qaNm3KmTNnuH79epIZQkNDsba2xsnJKc72nDlzYmOT8Jo0kZGRif6u0enerFLv/L8prm1tbbGxsTE2RJw5c4Y1a9Ywe/bsJHPVq1ePv/76i2fPniW5X2oz6WI/IiICC2nVF5lc/uIlmbZpN92+mYSFhRW6m3r82/vz+vLrN4NzgVp586qcUoi0Ya7VJlhoCdMQW/C+XYDZ2r7pmphcC25CXr58yaJFiyhbtiylSpVi5cqVtGjRgu7duyd77IoVK7C3t+err75i3rx5lClThjFjxjBy5Mg4+y1atIh+/fqRK1cupk+fTrVq1fj000+5d+9enP0MBgNNmzZl5syZNGnShAULFvDpp58yZ84c2rZt+8739vr1a7Zv306vXr3w9vamVKlSfPfdd8auJMkJCwtjxIgRjB49OtEPY2fPngWgbNmycbZ7enqSK1cu4/OJOX36NJGRkZQuXTpFmR49egS8+TAQKyIiIsGiPKXvi5o1axIaGkqvXr24evUqt2/fZvHixWzevJlRo0bF2//gwYPY2tpib29Pnjx54rX8FyhQAEdHR8aNG8ft27eZMWMGoaGhxnscMGAA/fr1M34zkZgyZcqgKArHjx9Pcr/UZtLdeCIjI2UmHmESzMzM+KTDF1Ss9wlLxo/g1OF9XOt2jWzlsoEWKufOrXZEIdKEQVHQak26XSpLi225PXbsGLVq1TJuP3r0KAD3799P0XkUReHPP/9k2bJlbNq0Cb1eT/Xq1VmxYgWtW7c2FonJWbt2bZwis3fv3vTu3ZsffviBSZMmYWVlRWRkJN9++y3lypXj4MGDxtbvEiVK0KVLF3LlyhXnfPv37+fIkSNUrVrVuL1YsWL07t2b48ePU7ly5SQz3bx5k127drFr1y4OHz5MREQE+fLlo0WLFjRu3JgaNWpgmcJaZ8KECdjY2DB48OBE93n48CHwphX8bTlz5uTBgwdJXsPf3x8AX1/fZPNERkYyd+5cfH1943T1KliwIEePHiUsLIxs2bIZt//1119A8u+LHj16cPnyZZYsWcLSpUuBN/+OLly4MN7A2xIlSlC1alUKFixISEgIK1asYNCgQTx48IBp06YBYGdnx6JFi+jWrRuzZ8/GzMyMadOmkTt3btauXUtgYCC///57sveb938Nc1euXEl0YHRaMOliPyIiQubYFybFxd2DUT+s5N+De5kz9EtCj4diY22BuZmZ2tGESBNS7Ju20qVLU6FCBaZNm4aXlxe1atXi6tWrfPnll1hYWBi7TyRl3rx5LFy4kMDAQLy9vRkyZAhdunQhX75875znv4V+WFgYERERVKtWjSVLluDv70/JkiU5deoUISEhfPfdd3G6ubRv3z5eEb1x40YKFy5MoUKFCA4ONm6vXbs2AIcOHUqy2J8xYwbDhw8HoFy5ckyePJlGjRol22c+IdeuXWPevHn8+uuvWCUxe1vsa57QPtbW1oSGhiZ5nZCQN7PHOadgVfd+/fpx5coVdu3aFee1/PLLL9mxYwdt27Zl8uTJ2NnZ8cMPP3Dq1Kk4GRNjZmZGvnz5aNCgAa1bt8ba2ppff/2V/v374+Hhwaeffmrcd/v27XGO7dq1Kw0bNmT27Nn079/f+OHts88+4+OPPyYgIABfX1/c3d0JDw9nxIgRTJ48GXt7e8aPH8/KlSuN//9296rY1+S/74X0YNLF/ps++9KyL0xPudr1+fnYRTqVKcDH+WTAmTBdBoNBin0T99tvv9G2bVvjoEUzMzO++uorjhw5QkBAQLLHz5kzh9u3b1OvXj2WL18eZ1Dou7p8+TLffPMNBw8ejFfUvnz5EoDbt28DxOuyYW5uTp63Fje8fv06V69eJXv27Ale78mTJ0nmyZ8/P/nz5ycwMJDz58/j6OiImZkZZmZm+Pn5vcutMXDgQCpXrkzLli2T3C/2A09CYwr0en2ifd7fpihKks/PmDGDn376iYkTJ/LJJ5/Eea5hw4YsWLCAkSNHGrvK5M+fn8mTJzN8+HDjdJ6JmTp1KvPmzeP69evGfdu0aUOtWrXo27cvjRs3TnDGHHgzScbgwYPZs2cPhw8fjjNw19nZmYoVKxr//N1335EjRw66du3Kzz//zOLFi1mzZg23bt2ibdu2XLlyJc77JPY1Se+xpCZd7JubmxMTk7I5YYXIbB7cvEGMwUDTwoXVjiJEmjEoCmbyzZVJ8/Ly4q+//uL69es8evQIPz8/PDw88PT0TNHsOT///DNLlixh69at5M6dm4YNG9K1a1caN26c4u4tAC9evKBGjRo4ODgwYcIE8uXLh7W1NWfOnGHEiBEYDIZ3vjeDwUDx4sUTHbjp7e2d5PHNmzenefPmBAQEsGvXLnbu3Mnw4cMZPHgwfn5+NGrUiEaNGlG9evUk7/XgwYPs3r2bzZs3c+vWLeP26OhodDodt27dwsXFBQcHB2P3nYcPH8bL9/DhQ8qXL59kZldXVwCeP38ep0vTf61YsYIRI0bQu3dvvvnmmwT36devH127duXChQtYWlpSqlQpli1bBpDs++KHH36gdu3a8T4UNG3alK+++opbt24l2b8+9r6TGkh769YtZs2axd69e9Fqtfz666/06tXL+K3NypUrWbduXZz7e/78ORB3fEJ6MOli38bGhkgVFi8QIj0c+G0tAFXfakkSwpRIN56sw8/Pz9hafeXKFR4+fEiXLl2SPa527drUrl2bkJAQVq1axbJly2jZsiVubm60b9+erl27UrJkyWTPc/jwYUJCQti8eXOcedyDgoLi7Jf7f2OkAgMD44wziI6O5tatW5QoUcK4LV++fJw/f546dep8UGtuwYIFKViwIF999RWhoaHs3buXnTt3snbtWubOnUuePHni5fyv2AXSWrRoEe+5+/fv4+vry5w5cxg0aBClSpUC3szF/9/C/sGDB9y7dy/O3PkJie1iFBQURPHixeM9v23bNrp3706LFi34/vvvkzyXnZ0dlSpVMv55//792NjYUKVKlSSPe/z4cYID+6OiogCSXRzs5s2bAIl+IwNvpidt2rSpcSzGgwcP8PT0ND7v6ekZb2xB7M+ocDo30pn0b1AbGxsiU2FqKyEyogsn/sLPzY3syXydKURmJt14sh6DwcDw4cOxtbWNN5gyKa6urgwePJhLly5x4sQJmjVrxrJlyyhVqhSlS5fmjz/+SPL42G+Q/tv9JDIyMt6UnWXLlsXV1ZWffvopTtG4Zs0aY8ttrDZt2nD//n1++umneNfT6XTvNa2sg4MDrVq1YsWKFTx69IiTJ0/Sv3//JI+pXbs2W7ZsiffInj07ZcuWZcuWLcb1DIoWLUqhQoX48ccf4xTMixYtQqPR0KpVqySvVaZMGSwtLY396//rzz//pF27dlSvXp01a9a809/t48ePs3nzZrp164ajo6Nx+8OHD/H39zcW8vCm5X/fvn3G8QPwZmGsDRs2kC1bNuN4jmfPnsX7UBAVFcXUqVOxtLSM82Huvw4dOsTvv//O9OnTjdvc3d2Ng5PhzVSib894dPr0aTQaTZwPMOnBpFv2ra2tidDrUBRF5toXJifk/l0a/a8FRghTFZOFW/bv3QzM1NdeuHAhL168MM7esmPHDuPUlP379zcWbAMHDkSv11OqVCmioqJYu3Yt//zzDytXrnyvBb0AKlasSMWKFZk7dy7r169n6dKl7N27l4YNGyZ6TOXKlXF2dqZz584MGDAAjUbDL7/8Eq/vuaWlJePGjaN///7Url2bNm3acOvWLVasWEG+fPni1BsdO3Zkw4YN9O7dm0OHDlGlShViYmLw9/dnw4YN7NmzJ94Ul/916tSpBIvmtyXXLcTHxyfB13LQoEG4u7vHGbAKb/rTN23alPr169OuXTsuXbrEwoUL6d69e7Kt0tbW1tSvX5/9+/czYcIE4/bbt2/TtGlT4weGjRs3xjmuRIkSxm9Fbt++TZs2bWjatCkeHh5cvnyZxYsXU6JECaZMmRLnuFGjRrFy5UqCgoKMYyZGjhxJhw4dqFChAj179sTGxoZff/2V06dPM2nSJCz+N3nL9u3bmTRpEq1atcLX15dnz56xdu1aLl26xJQpUxKcnjQmJoZBgwYxbNiwOK9pq1atGD58ONmzZ+f27dtcvHiRNWvWxDl23759VKlSxdjVKb2YdLEfO4gkOioSC8vER54LkdlcO3+aiMhI6cIjTJ7BYMhyffbd3NywtbVl3rB+quawtbX9oL7FM2fONA5mBdi8eTObN28GoEOHDsZi/6OPPmLu3LnGlt7y5ctz4MCBRFtV34W9vT3dunWjW7duybaiu7q6snPnToYMGcI333yDs7MzHTp0oE6dOsZVUmP169cPRVGYNWsWQ4cOpWTJkmzfvp0BAwZgbW1t3E+r1bJ161bmzJnDqlWr2LJlC7a2tuTNm5eBAwcm2/d8586djB8/Ptn7dHd3p1OnTsnul1KNGzdm8+bNjB8/nv79+5M9e3ZGjx7NmDFjUnT8F198QcuWLbl7966x/3tQUJBxkHPfvn3jHTN27FhjsR87dmDhwoU8e/YMLy8vBgwYwNdffx1nKs7EtG/fHjc3N7777jvjnPgFCxZk8eLF9OrVy7hf8eLFKVKkCKtXr+bp06fGsQEbNmygdevWCZ57yZIlPHv2jBEjRsTZ3rt3b4KCgpg9ezZ2dnYsX748ziJiL1++ZO/evSla3C21aZTkhktnYhs3bqRNmzas+ucqdg6OyR8gRCaxaMxw9m9YzY1hw3C1s1M7jhBpwmAw4DpxIkuWLKFHjx5qx0lVer2eoKAgfH194xSHse7cuZPu0/O9zc3N7b1b1rMig8FA9uzZadGiRYLddrKSmJgYihQpQps2bZg4caLacTKEuXPnMn36dG7cuJHiGY2Sk9zvkVhZomU/MiICKYeEKbn0918Uyp5dCn1h0sIiI1EUJU7/3KwisW4XImPQ6/VYWVnF6bKzatUqnj17Rs2aNdULlkGYmZkxYcIEvvzyS0aMGJHsVJmmLioqitmzZ/PNN9+kWqH/Lky62I/9lBMZITPyCNPy7OF9Pk3hUuRCZFah/5tNLSsW+yJj+/vvvxk8eDCtW7fG1dWVM2fOsGzZMooVK5Zo94+spm3btrRt21btGBmChYWFcUYkNZh0sR/76SlKZuQRJuTqqZNERkVRLQVLkQuRmYX+73e3k5OTukGEeEuePHnw9vZm/vz5PHv2DBcXFzp16mScxUWIjCRLFPuR+uSX2xYiszi4eT0aoMr/5noWwlRJy77IqPLkycP27dvVjiFEipj0fGbGJZ+l2Bcm5PI/xynq7oGTCv3+hEhPsS37UuwLIcT7M+liP3bls9AkljsWIjMxGAw8f/SAGnmlC48wfS+lZV8IIT6YSRf7Li4umJmZ8Tz4idpRhEgVl/4+RmR0NNVkfn2RBYTq9Zibm6sye4UQQpgKky72tVotOXK48zL4qdpRhEgVh7dtfLPUtvTXF1lAaEQEjtmyyQroQgjxAUy62Adw9/DghRT7wkRc+fdvSnh44JjE4hlCmIoXOh1O0oVHCCE+iMkX+x4e7tKNR5gEg8HAiyePqJk3r9pRhEgXj8LC8MqVS+0YQgiRqZl8sZ/Tw4PQEHWXHBciNZz76zBR0l9fZCH3X70il6wiK4QQH8Tki313d+mzL0zDkW2b0Go0VJTiR2QRD8LCyCUt+0JkStOnT6dQoUIYDAa1o2QIFStWZPjw4apc2+SLfQ8PD54HP0VRFLWjCPFBrp7+h1KenthbWakdRYg0pygKD16+zLLF/p07dzhz5oyqjzt37qTa/UyePBmNRkOxYsUSfP748eNUrVoVW1tbPDw8GDBgAK9evUq164v/t379eipVqoSdnR1OTk5UrlyZgwcPxttv2bJlFC5cGGtra/z8/FiwYEGKrxEaGsq0adMYMWIEWu3/l5p58uRBo9HEe/Tu3TvO8Q8fPmTkyJHUqlWLbP8bpH/48OEUX3/z5s20bduWvHnzYmtrS8GCBRkyZAgvXrxI8PXo0KEDfn5+aDQaatasmeA579+/T6NGjXBwcKBIkSLs2LEjwevmyJGDly9fxntuxIgRfP/99zx69CjF95FaTHoFXXhT7Efodehev8LWPpvacYR4LwaDgdCnj6lZqZLaUYRIF8/Cw4mIisqSxf6dO3coXKgQ4Tp1F4S0tbHhqr8/Ph/4beK9e/eYMmUKdnZ2CT5/7tw56tSpQ+HChZk9ezb37t1j5syZXL9+nT/++OODri3iGjduHBMmTKBVq1Z06dKFqKgoLl26xP379+Pst2TJEnr37k3Lli356quvOHr0KAMGDCA8PJwRI0Yke52ff/6Z6OhoPvvss3jPlSpViiFDhsTZVqBAgTh/DggIYNq0afj5+VG8eHFOnDjxTvfZs2dPPD096dChAz4+Ply8eJGFCxfy+++/c+bMmTjT+S5atIjTp09Trlw5QkJCEj1n586duX//PtOmTePYsWO0bt0af39/8vyva61er2fo0KFMmjQpwbVBmjVrhoODAz/88AMTJkx4p/v5UCZf7Lu7uwPw4ukTKfZFpnXq0D6iYmKo5iuLaYms4X5oKECWLPaDg4MJ1+n4sUULCri5qZLhWnAwPTdvJjg4+IOL/aFDh1KxYkViYmIIDo4/hm706NE4Oztz+PBhHBwcgDctwD169GDv3r3Ur1//g64PEBYWRkhIiLEwy4r+/vtvJkyYwKxZsxg8eHCi++l0Or7++msaNWrEpk2bAOjRowcGg4GJEyfSs2dPnJ2dk7zW8uXLadq0KdYJzBzn5eVFhw4dkjy+TJkyhISE4OLiwqZNm2jdunUK7vD/bdq0KV4LfZkyZejcuTNr1qyhe/fuxu2//PILXl5eaLXaRL950ul0HDx4kMOHD1O9enV69+7N8ePH2bNnD7169QJg5syZODo6xjn3f2m1Wlq1asWqVasYP358uk4pbPLdeDw9PQEIefxQ5SRCvL8/d/yGmVZLBW9vtaMIkS4eZOFiP1YBNzdKeXqq8kitDxl//vknmzZtYu7cuQk+Hxoayr59++jQoYOx0Afo1KkT9vb2bNiw4YOu/9dff9G1a1dy5szJ1q1bk91/+fLl1K5dmxw5cmBlZUWRIkVYtGhRvP0MBgPjxo3D09MTW1tbatWqxZUrV8iTJw9dunSJs++LFy8YNGgQ3t7eWFlZkT9/fqZNm/bBfdkNBgN///033377LT169Eh2/7lz5+Lh4cHAgQNRFCXRblKHDh0iJCSEPn36xNnet29fXr9+za5du5K8TlBQEBcuXKBu3bqJ7hMZGcnr168TfT5btmy4uLgkeZ2kJNQVp3nz5gBcvXo1znZvb+84XY0SotfrURTF+CFHo9Hg5OREeHg48KaLz9SpU5k3b16S56pXrx63b9/m3Llz73A3H87ki/08efJgZmbGg1s31Y4ixHsLOHOKMl5e2Fpaqh1FiHTxIDQUc3NzcuTIoXYU8Z5iYmLo378/3bt3p3jx4gnuc/HiRaKjoylbtmyc7ZaWlpQqVYqzZ8++83UfP37MjBkzKFSoENWqVWPnzp10796dTz/9NNljFy1aRO7cuRk9ejSzZs3C29ubPn368P3338fZb9SoUYwfP56yZcsyY8YM/Pz8aNCgQbwCNjw8nBo1arB69Wo6derE/PnzqVKlCqNGjeKrr75653t78eIF69evp1OnTri7u1OpUiV++OEH3FLw4ezAgQOUK1eO+fPnkz17drJly0bOnDlZuHBhnP1iX/O3fyZlypRBq9Um+zM5fvw4AKVLl07w+YMHD2Jra4u9vT158uRh3rx5yWZPDbF95VPyWr3N2dmZfPnyMWXKFIKCglizZg3nzp2jfPnyAAwfPpyGDRtSvXr1JM9TpkwZAI4dO/bOGT6EyXfjsbS0JI9vXh4E3VA7ihDvJTo6mtCQp9SoUkXtKEKkm/uhoeR0d8fMzEztKOI9LV68mNu3b7N///5E93n48M237jlz5oz3XM6cOTl69GiKrhUTE8Mff/zBsmXL2LlzJ4qi8PHHHzN58mSaNm2KhYVFis5z5MiROP25+/Xrx8cff8zs2bPp27cv8ObDxOzZs/n000/ZsmWLcd/x48czbty4OOebPXs2N27c4OzZs/j5+QHQq1cvPD09mTFjBkOGDME7mW9sL126xK5du/j99985fvw40dHRFCtWjO7du9OoUSMqVaqU7N+T58+fExwczLFjxzh48CBjx47Fx8eH5cuX079/fywsLIzdUR4+fIiZmVm8D9qWlpa4urry4MGDJK/l7+8PgG8C3U5LlChB1apVKViwICEhIaxYsYJBgwbx4MEDpk2bluR5P9S0adMwMzOjVatW73X8jz/+SKtWrVi3bh0AgwYNokqVKhw/fpwtW7bE+8YgIV5eXlhaWnLlypX3yvC+TL5lH6BQoYLSsi8yrX8O7CZa+uuLLOZ6cDB+bw3aE5lHSEgIY8aM4dtvvyV79uyJ7qf73yBkqwRmGbO2tjY+n5Svv/4aHx8fmjRpwtWrV5k0aRJ3795l586dtGzZMsWFPhCn0H/58iXBwcHUqFGDmzdvGmdYOXDgANHR0fG6ufTv3z/e+TZu3Ei1atVwdnYmODjY+Khbty4xMTH8+eefSebp27cvxYsXZ+TIkZibmzNv3jxu377NxYsX+e6776hatWqKPhDHdtkJCQlh6dKlDB06lDZt2rBr1y6KFCnCpEmTjPvqdDosE/kWOSU/k5CQEMzNzf+vvTuPqyn//wD+uu37Ym2kEkZkGbuxJGTfxz7G3ozsMmMfxtiyMzN8MSPFpGwRZReSJZWlZGmhFFFU0r7d7u8P0/1J0aI6de/r+Xjcx0y3e899XVGvc3qfz4GWllaBz7m7u2PhwoUYMmQIpkyZgqtXr6JPnz7SE7PLi4uLC/bu3YtffvlFutNVUj169EBUVBRu3bqFqKgobNu2Dbm5uZgzZw5++eUXmJiYYNeuXWjcuDHMzMywe/fuQreT93ehIsn8kX0AaGxmhnuux4SOQVQq1zzcoKyggPZyPLtM8if07VtY9eghdAwqpWXLlqFatWqFFuAP5ZXrzMzMAp/LyMjIV74/xc7ODgDwww8/YPv27UWePPo5N27cwIoVK+Dj4yOdx87z7t076OrqIjIyEgDQsGHDfJ+vVq1agdcOCwvD/fv3P7nD8/r168/madasGQwNDREdHY1bt25BU1MTioqK6N+/f5G/EfhQ3p+jsrJyviPbCgoKGD16NFasWIGoqCgYGxtDXV0dWVlZhW6nuF+T4hKJRJg3bx7Onz8PLy+vIk/cLY1r167B2toaffr0wdq1a79oW1paWujQoYP0Y0dHR8TExGDx4sXw9PTEggULcODAAYhEIowdOxZmZmbo3r17vm1IJJIKPTkXkJOyb2ZmhpjnUcjOyoSyCtcop6olLOAO2tatC7USHJ0iqspyxGI8ffMGs83NhY5CpRAWFoZ//vkHf/zxR76Rj4yMDGRnZ+PZs2fQ0dFBtWrVpOM7eeM8H3r16pV0kY3POXbsGOzt7XHw4EEcO3YMQ4cOxeTJk9GzZ88iT7z80NOnT2FlZYXGjRtj69atMDIygoqKCs6cOSM9iltSubm56NWr1ycvpvTxkpMfmz59OqZPn46AgACcOnUKp0+fxowZM5Cbm4sWLVpgwIABGDBgAL799tvPHuGvVq0a1NTUoKenV+BxeeM6b9++hbGxMb766iuIxWK8fv063yhPVlYW4uPji/yaVK9eHTk5OUhOToa2dtGrIObttCQkJBT52JIKDAzE4MGD0axZM7i6ukJJqexqb1JSEn799Vds3rwZmpqaOHjwIEaMGCE9N2TEiBFwdnYuUPYTExNLdd7Al5CLMZ5GjRohNzcXMVGRQkchKpHs7GwkJ8TBsn59oaMQVZiIt2+RLRajSZMmQkehUoiOjpaON5iamkpvvr6+CA0NhampqXSd8WbNmkFJSQm3b9/Ot42srCwEBASgZcuWRb7esGHDcObMGURGRmLp0qXw9fVFnz59YGJigmXLluHJkyfFyu3h4YHMzEy4u7vDxsYG/fv3R8+ePQscyTYxMQGAAtuNj4/H27dv893XoEEDpKSkoGfPnoXeirusacuWLbFs2TL4+PggNjYW+/fvR+PGjbFr1y506dLlsyvfAO+P4Lds2RJv3rwpcNQ+b4cs77cPeX/mH39Nbt++jdzc3CK/Jo0bNwbwflWe4ggPD8/3+mXl6dOn6Nu3L2rVqoUzZ84UOlb0JVatWgVTU1P88MMPAN7/OX64I1SnTp0C1y+Ijo5GVlZWhX9vk4uyb2ZmBgA8SZeqHJ9zHsjJzYWFHK8NTfIn+M0bAGDZr6KaNWsGNze3AremTZvC2NgYbm5usLa2BgDo6uqiZ8+eOHDgAJKTk6XbcHJyQkpKSonWV69bty6WL1+Op0+f4sKFC+jcuTM2b96Mr7/+Gl27doWfn99nn593xFsikUjve/fuHRwdHfM9zsrKCkpKSgWW5Px4VRsAGDVqFHx8fHD+/PkCn0tMTEROTk6x31+eGjVqYMKECTh8+DDevHmDK1euFHrxqo+NHj0aYrEY+/fvl96XkZEBZ2dnmJubS4tqjx49UK1atQLvb9euXdDQ0MCAAQM++zod/7v448c7CwkJCRCLxfnuy87Oxvr166GiolLgCHhxRUVFSU8KzhMTE4PevXtDQUEB58+fL/MdidDQUOzYsQN//vmndCSndu3a+XI8fvwYBgYG+Z53584dAECnTp3KNE9R5GKMp3bt2tDW0cHLZyz7VLXcOOMOFUVFtOW8PsmR0DdvoKejI70oIlUtNWrUKHSZy7y19j/+3Nq1a9GpUydYWlpi6tSpePHiBbZs2YLevXujb9++JX59kUiEXr16oVevXoiPj4eTkxP27t2LmzdvSpdKLEzv3r2hoqKCQYMGwcbGBikpKdizZw9q1aqVb8yodu3amDt3LrZs2YLBgwejb9++CAwMxNmzZ1GjRo1889gLFiyAu7s7Bg4ciEmTJqFNmzZITU1FUFAQXF1d8ezZs8+OdFy5cgUhISFFvufi/FuxsbGBvb09Zs6cidDQUBgbG8PJyQmRkZHw8PCQPk5dXR2rV6/GzJkzMXLkSPTp0wfXrl3DgQMHsHbt2iLXv69fvz6aNWsGT09PTJkyRXq/u7s71qxZgxEjRsDU1BQJCQlwcXHBgwcPYGdnV6AY5500/PDhQwDvdwCvX78O4P05IXkmTJiAq1ev5ttJ69u3L8LDw7Fw4UJcv35d+ry8P6tevXpJP/b29paeKP3mzRukpqZKX7tr166FLqc5b948jB49Ot/fpxEjRmDIkCFYunQpgPe/KTp16lS+5128eBHGxsZo1arVZ/8My5pclH2RSIRGjRrxyD5VOU/u30UHY2OoluGcIVFlFxIXB3Nz8wo/ia2yCa3gFTuEeu3WrVvD09MTixYtwrx586CtrQ1ra2usW7fui7ddvXp12NrawtbW9rMXcQLeTwG4urpi2bJlmD9/PgwMDDB9+nTUrFkzX2kF3i/jqKGhgT179sDT0xMdO3bEhQsX0KVLl3xXjdXQ0MDVq1dhZ2eHo0eP4t9//4WOjg4aNWqElStXQldX97OZ9u/fn+9I/Kd06NBBetGoT1FXV8fly5excOFCODg4IDU1FS1btsTp06fRp0+ffI+dMWMGlJWVsWXLFri7u8PIyAjbtm3D3Llzi8wCAFOmTMFvv/2G9PR06RhU8+bNYW5ujgMHDuDNmzfSaykcOXKk0N/gLF++PN/HDg4O0v//sOwXJjAwEACwcePGAp+ztLTMV/YvX76MlStXFvraK1asKFD2z5w5A29vb4SGhua7f+DAgVi7di22b98OiUSCdevWoV+/ftLP5+bm4tixY7C2tq7w720iyYe7QjJs/PjxuP0oBGtdTgodhahYsjIyMK51Qyzp1g3zi7hQB5EssdyzB2379oW9vb3QUcpNRkYGIiIiYGpqmq8cAu/HEpo0boy0Yiw7WZ401NXxODi42HPl8i4xMRH6+vpYs2YNfv31V6HjCOrdu3eoX78+Nm7cKB3ZkncnTpzA2LFj8fTp00KvK1Ean/s+8iG5OVzYrFkzHHNzg1gs5kVaqEq4cdYdYs7rk5zJzc1F2Js3GC/H8/rGxsZ4HBxc4Wtxf6xGjRos+p/w4RHrPHljSt26dav4QJWMrq4uFi5ciE2bNmHy5MklWhVJVm3YsAGzZs0qs6JfEnJT9tu3b4/01FREhz+B8ddmQschKtLNs+5QVVJCq2IsPUckK8ITEpCWlYUWLVoIHUVQxsbGLNqV2OHDh7Fv3z70798fWlpauH79Og4ePIjevXujM692DgBYtGgRFi1aJHSMSsPHx0ew15abst+2bVuIRCI8uX+PZZ+qhKdBgehobAwVzuuTHLn33zKAbdu2FTgJ0ae1aNECSkpK2LhxI5KSkqQn7X54JVqiykJuWoS2tjYaNzFH2P176DF8jNBxiD4rIy0NKYkJsGzdUugoRBXq7suXaGBq+kVXQSUqb3knFRNVBXI1RPVth/Z4GhQgdAyiIl0/7QaxRAILU1OhoxBVqICYGLT9zPKIRERUMnJV9jt06IBnIY+QmSHsCgdERbl57hQ0lJXRUoATeYiEkiMWI/DlS47wEBGVIbkq++3bt4dYLEbEowdCRyH6rIiH99HJxARKXDmK5MjD16+RlpWFb7/9VugoREQyQ67KfrNmzaCmpoaw+/eEjkL0SempKUh5lwhLjvCQnPF//hxKSkpo06aN0FGIiGSGXJV9ZWVltG7dhmWfKjWvk0eRK5GgC9fXJznj9+IFWn3zTYH1y4mIqPTkquwDQIcO7RH+IEDoGESf5HvhLDRVVNCC8/okZ/xfvsS3nToJHYOISKbIYdnvgFdRkUiMeyN0FKJCPXschC716kGRVxwkORL97h0i4uLQpUsXoaMQEckUuWsTeZexvu9zTdggRIVITU5CalISunJen+TM1YgIiEQi9OjRQ+goRJXas2fPIBKJsG/fPul9v//+O0QiUam2V69ePQwcOLDIx3l5eUEkEsHLy6tUr/Mpfn5+UFFRQWRkZJlut6oaM2YMRo0aVabblJuLauWpXbs2vvmmJQKvX0XXQcOEjkOUzxW3I8iVSGDBeX2SM17h4Wj1zTeoUaOG0FEqhaioKMTFxQmaoUaNGjA2NhY0A8m+X3/9Fd9//z1MTEyk9/n5+WHfvn3w9fXF/fv3kZOTA4lE8sltxMbG4rfffsOpU6cQHx8PAwMDWFlZYe/evcXKcPfuXfz++++4fv06MjIyUL9+fUydOhVz5swBAKSlpcHR0REnT55EUFAQUlJS0LBhQ0ydOhVTp06F4gcr5yUmJmLmzJk4c+YM9PX18euvv8La2jrf692+fRtdu3bFw4cPYfrRwb1Fixahbdu2CAwMxDfffFOs/EWRu7IPAH369Ia94z5IJJJS7wkTlQffi2ego6qKZrVrCx2FqMJIJBJcffYME6dNEzpKpRAVFQWzJmbISMsQNIeahhpCHoew8FcyJiYmSE9Ph7KystBRvlhAQAA8PT1x8+bNfPefOXMG9vb2aNGiBerXr4/Q0NBPbuP58+fo3LkzAGDatGkwNDTEy5cv4efnV6wMFy5cwKBBg9CqVSssX74cWlpaePr0KV68eCF9THh4OGbPng0rKyv8/PPP0NHRwfnz5zFjxgzcunUL+/fvlz52/vz58PLywsqVK/HkyRP89NNPaNKkCTr9dz6SRCLBnDlzYGtrW6DoA0CrVq3Qtm1bbNmyBf/++2+x3kNR5LLs9+7dGxs3bkRUaDBMzJoIHYdI6nnwI3Q1NYUC5/VJjjx+/RqxSUno2bOn0FEqhbi4OGSkZaDe6npQM1UTJENGRAaeLX+GuLg4lv1KRiQSQU1NmL8XZc3R0RHGxsYFrq0xffp0LFq0COrq6pg1a9Zny76NjQ2UlJTg7++P6tWrl+j1k5KSMGHCBAwYMACurq6f/NlrYGCAoKAgNG3aNN/rTpkyBY6Ojli+fDkaNmwIADh16hQ2btyICRMmAADu378PDw8Padl3dnZGZGQkli5d+slco0aNwooVK7Bz505oaWmV6D0VRi4bRefOnaGuro7AG1eFjkIklZz4FikpyRzhIbnjFR4OVRUVnpz7ETVTNWg00RDk9qU7GZGRkZgxYwbMzMygrq6O6tWrY+TIkXj27FmBx96/fx+WlpZQV1dH3bp1sWbNGjg6OkIkEhV4/NmzZ2FhYQFNTU1oa2tjwIABePjwYZF5PjXTvm/fvgKvkzfDfv36dbRv3x5qamqoX79+gaOsCQkJmD9/Ppo3bw4tLS3o6OigX79+CAwMlD4mNjYWSkpKWLlyZYHXDgkJgUgkwo4dO4q9PaDwmf3CODo6okePHqhVqxZUVVVhbm6OXbt2ffLxFy5cQMuWLaGmpgZzc3McP378s9vP4+vri759+0JXVxcaGhqwtLTEjRs3ivXcEydOoEePHgW+NrVr1y7WErzBwcE4e/YsFixYgOrVqyMjIwPZ2dnFem0AcHFxQWxsLNauXQsFBQWkpqYiNze3wONq1KiRr+jn+e677wAAjx8/lt6Xnp4OfX196cfVqlVDWloaACA1NRWLFy/GunXrPlvie/XqhdTUVFy8eLHY7+Vz5LLsq6mpoWtXSwTeZNmnyuPSsYOQSCSw4Mm5JGeuRkSgS5cuXF9fhvj7++PmzZsYM2YM/vrrL0ybNg2XLl1Ct27dpMUHAKKjo9G9e3c8fPgQS5Yswbx58+Ds7Iw///yzwDadnJwwYMAAaGlpYcOGDVi+fDkePXqELl26FLoT8SWePHmCESNGoFevXtiyZQv09fUxadKkfDsW4eHhOHHiBAYOHIitW7diwYIFCAoKgqWlJV6+fAngfWm1tLTEkSNHCrzG4cOHoaioiJEjRxZ7eyWxa9cumJiYYOnSpdiyZQuMjIwwY8YM/O9//yvw2LCwMIwePRr9+vXDunXroKSkhJEjRxZZNi9fvoyuXbsiKSkJK1asgJ2dHRITE9GjR48ix2iio6MRFRWF1q1bl/i95fH09ATw/s/ZysoK6urqUFdXR79+/Yr1d8LT0xM6OjqIjo6GmZmZdCdr+vTpyMgoeowuJiYGAPKda9SuXTts3boVYWFhOH/+PM6dO4f27dsDAOzs7GBoaIjx48d/drvm5uZQV1cv9k5TUeRyjAd4P7e/eMlSZGakQ1WNP2BIeP6XLkBXTQ1NatYUOgpRhckWi3E9MhLLJk8WOgqVoQEDBmDEiBH57hs0aBA6duyIY8eOScvOhg0b8PbtW9y9exctW7YEAEyePBlff/11vuempKRgzpw5+PHHH/HPP/9I7584cSLMzMxgZ2eX7/4vFRISAm9vb1hYWAB4P1ZhZGQER0dHbN68GQDQvHlzhIaG5hv9GD9+PBo3boy9e/di+fLlAIDRo0fDxsYGDx48QLNmzaSPPXz4MCwtLVH7v3O0iru94rp69Wq+HehZs2ahb9++2Lp1K2bOnJnvsaGhoTh27BiGDXu/cIm1tTUaN26MRYsWoVevXoVuXyKRYNq0aejevTvOnj0rPTpvY2ODpk2bYtmyZbhw4cIn8wUHBwNAoXPrxRUWFgYAmDp1Ktq1a4fDhw8jKioKK1euRM+ePXH//n1oaGh89vk5OTkYMmQIrK2tsW7dOnh5eWH79u1ITEzEwYMHP/ncrKws/PHHHzA1NUW7du2k92/btg39+/dHo0aNAADDhw/H999/j4iICGzbtg2XL18u8nxRJSUlGBkZ4dGjRyX54/gkuTyyD7yf28/KzMDj275CRyECADwPfQzL+vU5r09yxf/FC6RmZn6yUFDV9GHJzM7ORnx8PBo2bAg9PT3cvXtX+rlz586hY8eO0qIPvB97+OGHH/Jt7+LFi0hMTMT333+PuLg46U1RUREdOnTAlStXyjS/ubm5tOgDQM2aNWFmZobw8HDpfaqqqtLv12KxGPHx8dDS0oKZmVm+9zhs2DAoKSnh8OHD0vsePHiAR48eYfTo0SXeXnF9+DV49+4d4uLiYGlpifDwcLx79y7fY+vUqSMdSQEAHR0dTJgwAffu3ZMevf5YQEAAwsLCMHbsWMTHx0u/JqmpqbCysoK3t3ehIzF54uPjASDfyEtJpaSkAHg/U3/69GmMGjUK8+fPx549e/D06VO4uLgU+fy0tDRMmDABf/31F4YNG4a//voLNjY2OHTokHRnojCzZs3Co0ePsGPHDigp/f+x8+bNmyMsLAz+/v4ICwuTngvwyy+/YPjw4fj2229x/PhxfPPNNzA1NcWqVasKXWlIX1+/zFbkkttWYW5ujjqGhpzbp0ohMT4OaSnJ6Mp5fZIz50JCULN69Xxlj6q+9PR0/PbbbzAyMoKqqipq1KiBmjVrIjExMV/RjIyMlJ7Y+KGP78srXT169EDNmjXz3S5cuIDXr1+Xaf7CTkrW19fH27dvpR/n5uZi27Zt+Prrr/O9x/v37+d7jzVq1ICVlVW+UZ7Dhw9DSUlJeiS9JNsrrhs3bqBnz57Q1NSEnp4eatasKT0p9OPtNWzYsMDR5rwj058ah8n7mkycOLHA18Te3h6ZmZnFyv25JTWLkrdDM2rUqHwHykaOHAklJaUCq/x86vnff/99vvvHjh0LAPDx8Sn0eZs2bcKePXuwevVq9O/fv8Dn1dTU0LZtW+nf48uXL+PChQtYv349QkJCMGbMGNja2sLBwQE7d+4s9PyLslwxUm7HeEQiEfr07o3LLPtUCVxydYEE4Lw+yRWJRAKP0FAMHTYs3zrVVPXNnj0bjo6OsLW1RceOHaGrqwuRSIQxY8Z89mjvp+Q9x8nJCQYGBgU+/+GR1cJ8qjSJxeJC7//U38cPi6mdnR2WL1+OKVOmYPXq1ahWrRoUFBRga2tb4D2OGTMGkydPRkBAAFq2bIkjR47Aysoq36x3SbZXlKdPn8LKygqNGzfG1q1bYWRkBBUVFZw5cwbbtm0r1dfgY3nb2LRp0yd31j93Emreyjkf7kCVVJ06dQBAOgqVR1FREdWrVy9y23Xq1MHDhw8LPL9WrVqfzLZv3z4sWrQI06ZNw7Jly4rMKBaLMXfuXCxevBiGhoZYvXo1OnXqhMn/jS7a2NjA2dlZ+nGet2/fFhhnKy25LfsA0L9/fzg6OuJVZAS+MmHJIuHcvnwB1TU00IgXFCI58jA2FhFxcRg+fLjQUaiMubq6YuLEidiyZYv0voyMDCQmJuZ7nImJCZ48eVLg+R/f16BBAwDvS1hplmjNGxVJTEyEnp6e9P4vuWqrq6srunfvXuDCTYmJiQUuDjd06FDY2NhIR3lCQ0OxZMmSUm+vKB4eHsjMzIS7u3u+31J8atzpyZMnBY4k5y13We8Tv3HO+5ro6OiU6mvSuHFjAEBERESJn5unTZs2AN6f7PuhrKwsxMXFoWYR58C1adMGFy9elJ6gmyfvhOiPn3/y5En8+OOPGDZsWKEnOhdm165dSE5Oxvz586XbzttJAd7vcHycPycnB8+fP8fgwYOL9RpFkdsxHuB92dfQ1MTNsx5CRyE5F/0kBJb16/MibyRXPB4/hq6ODrp37y50FCpjioqKBcYztm/fXuBIep8+feDj44OAgADpfQkJCXB2di7wOB0dHdjZ2RW6tOKbN28+myevmHp7e0vvS01NzXcxpJIq7D0ePXq0QHEDAD09PfTp0wdHjhzBoUOHoKKigqFDh5Z6e8XJBuT/TcS7d+/g6OhY6ONfvnwJNzc36cdJSUn4999/0bJly0J/kwK8L8oNGjTA5s2bpbPzHyrqa2JoaAgjIyPcvn27yPfzKd26dUOtWrXg7Oycb/Wcffv2QSwW5zsXKC4uDsHBwflWgxo1ahQAFNjBsre3h5KSErp16ya9z9vbG2PGjEHXrl3h7OxcrPPrEhISsGLFCmzatEl6bYTatWtLT04G3i/b+fGf8aNHj5CRkSFdm/9LyfWRfQ0NDQweNAg3z7lj+LQ5QschOZUQG4O01FSur09y51RoKAYNHgwVFRWho1AZGzhwIJycnKCrqwtzc3P4+PjA09OzwEWPFi5ciAMHDqBXr16YPXs2NDU1YW9vD2NjYyQkJEgPgOjo6GDXrl0YP348WrdujTFjxqBmzZqIiorC6dOn0blzZ+l69YXp3bs3jI2NYW1tjQULFkBRUREODg7SbZT2Pa5atQqTJ09Gp06dEBQUBGdnZ9SvX7/Qx48ePRrjxo3Dzp070adPn3y/YSjN9j6nd+/eUFFRwaBBg2BjY4OUlBTs2bMHtWrVwqtXrwo8vlGjRrC2toa/vz9q164NBwcHxMbGfnLnAAAUFBRgb2+Pfv36oWnTppg8eTIMDQ0RHR2NK1euQEdHBx4enz+YOmTIELi5uRX4rUJkZCScnJwAQLozsGbNGgDvfxuUt5qTqqoqNm3ahIkTJ6Jr164YP348oqKi8Oeff8LCwiLfORE7duzAypUrceXKFWmJb9WqFaZMmQIHBwfk5OTA0tISXl5eOHr0KJYsWSI9Ah8ZGYnBgwdDJBJhxIgROHr0aL730aJFC7Ro0aLA+1u+fDmaN28uXV4VeL86z6pVqzB9+nSYmJjg77//xtatW/M97+LFi9DQ0CizhQvkuuwD7/fqDh06hBfhYahbv2xmo4hKwjNvXp9ln+TI0/h4PHz1Cqs/+GFM+WVEFL3Od2V97T///BOKiorSI66dO3eGp6cn+vTpk+9xRkZGuHLlCubMmQM7OzvUrFkTM2fOhKamJubMmZPvSrFjx45FnTp1sH79emzatAmZmZkwNDSEhYVFgXnnjykrK8PNzQ0zZszA8uXLYWBgAFtbW+jr6xf53E9ZunQpUlNT4eLigsOHD6N169Y4ffo0Fi9eXOjjBw8eDHV1dSQnJ+dbhae02/scMzMzuLq6YtmyZZg/fz4MDAwwffp01KxZE1OmTCnw+K+//hrbt2/HggULEBISAlNTUxw+fLjA1+tj3bp1g4+PD1avXo0dO3YgJSUFBgYG6NChA2xsbIrMOWXKFOzYsQM3btzId1G9iIiIAkuN5n1saWmZb536CRMmQEVFBevXr8eCBQugp6cHGxsb2NnZFetcoN27d8PY2BiOjo5wc3ODiYkJtm3bBltb23x58k42/njZUgBYsWJFgbIfFBQEe3t7+PrmX/WxefPmcHR0xO+//47k5GTMmDEDU6dOzfeYo0ePYtiwYdDW1i4yf3GIJF9yGrQMyMjIQI2aNTFw8jSMmvmz0HFIDi0a2Q/vwp8gdP58jvGQ3Pjj+nVsuH4dcfHxn10HW1ZlZGQgIiICpqam+QotAERFRcGsiRky0oQr+wCgpqGGkMchha5MU95sbW3x999/IyUlhSdvyzgrKyvUqVNHeiRf3gUEBKB169b5rj3xKZ/7PvIhuT+yr6amhqFDhuD6OQ+WfRLEy6dhGNCgAYs+yRWPkBD069dPLot+UYyNjRHyOKTM1tgurRo1alRI0U9PT8+3Jnx8fDycnJzQpUsXFn05YGdnBwsLC6xZswYmJiZCxxHc+vXrMWLEiDJdjljuyz7wfo7O2dkZUWEhMP7arOgnEJWRN6+ikZaWxiU3Sa68ePcOd54/x7x164SOUmkZGxsLckRdCB07dkS3bt3QpEkTxMbGYu/evUhKSirxFWOpaurQoQOysrKEjlFpHDp0qMy3ybKP9yey6Ojq4sbZkzD+eqHQcUiOXDx8AAB4MS2SK4cDA6GupoZBgwYJHYUqgf79+8PV1RX//PMPRCIRWrdujb1796Jr165CRyOSCXK99GYeVVVVfDd0KG6dO/VFV3IjKqm73pdhoK0Nky+4XDhRVSKRSOASFIThI0ZAR0dH6DhUCdjZ2SE0NBRpaWlITU3FtWvXSrVuOxEVjmX/P6NHj8aL8CeIDHkkdBSSI7ERT9CN6+uTHPF9/hxP37wp9QooRERUMiz7/7GyskK16tVx9aSr0FFITryKjEBaejqX3CS54hIQABMjo3wXqyEiovLDsv8fFRUVTBg/Ht7ux5DNE0WoAni6ugAAT84luZGalQW3R48wcfLkYl19koiIvhy/237gp59+QmJ8HPwvnxc6CsmBgGtXYKijA+OPrqJIJKs8Hj9GckYGJk6cKHSUSoPniRFRaRX3+wfL/gfMzc3RsWMnXPrviCtReYp9Fo7uDRoIHYOowrgEBsKya1fUr19f6CiCU1ZWhkgkQmpqqtBRiKiKSk1NhUgkgrKy8mcfx6U3P/LTTz/C2toar188R626RkLHIRn1/GkY0jMyOK9PciPy7Vt4h4dj32+/CR2lUlBUVISuri7evHmDzMxM6OjoQElJiSfrE9FnSSQS5OTkICkpCUlJSdDT0yvy4nMs+x8ZNWoU5tra4uJRZ/wwb7HQcUhGXTr6/rdHXVj2SU4cuHcPWpqaGD58uNBRKg0DAwOoq6vj9evXSEpKEjoOEVUhioqK+Oqrr6Crq1vkY1n2P6KpqYlJEyfCycUFo2bOg7KKqtCRSAYF3vCCsZ4eDIvxj5SoqsvIzobj3buYOGkStLS0hI5TaYhEIujp6UFXVxdisRg5OTlCRyKiKkBJSQmKiorF/k0gy34hZsyYge3bt+PW+dOwGDRM6Dgkg15HRWJ0U3OhYxBViGMPHiAuJQVz5swROkqlJBKJoKSkBCUl/kgmorLHE3QL0bhxY3Tv0QPnDu4XOgrJoMiQx8jIzOCSmyQXJBIJdvv5oX+/fmjUqJHQcYiI5A7L/ifMmjkTwXf9EfH4gdBRSMbkrfbEeX2SBzciIxH06hVs580TOgoRkVxi2f+EwYMHw7BuXZzaby90FJIxgTe9Ub9aNRhoawsdhajc7fL1hXnjxujZs6fQUYiI5BLL/icoKSnhl59/xrVTx/E6+oXQcUiGxD2PQneuM05y4FlCAs4EB2PuvHlcUpKISCAs+5/x008/QVdXF+4Ou4SOQjIi/GEQMrIyOa9PcuEfPz/o6+pi3LhxQkchIpJbLPufoaWlBdu5c3HJ9SAS494IHYdkgOd/8/qdTUwETkJUvpIzM3EgMBA206dDQ0ND6DhERHKLZb8Is2bNgrKyEk7/y9l9+nJBt67j6xo1UJNrjZOMs/fzQ0ZODmbOnCl0FCIiucayXwR9fX3MmD4d5w/uQ2rSO6HjUBUX/+I5unFen2RcSmYmtt+6Besff4ShoaHQcYiI5BrLfjHMmzcPOVlZOOfCdfep9EID7iAzOwsWXHKTZJy9vz+Ss7KwePFioaMQEck9lv1i+OqrrzBlyhSccbJHZnqa0HGoirp07BBE4Pr6JNvyjupPmTIFxsbGQschIpJ7LPvFtGDBAiQnvsUl14NCR6Eq6oHvDTSuVQvVeLIiyTCH27eRlJGBpUuXCh2FiIjAsl9spqam+P777+HusBvZWVlCx6Eq6O2raM7rk0xLzcrCXz4+mDR5Mo/qExFVEiz7JbBkyRLEx77CxSMHhI5CVcxD/1vIzM7mCA/JNAd/fyTyqD4RUaXCsl8C5ubmmDhxIlx3bkNqcpLQcagKuXL8/bw+19cnWZWWlYW/bt3CpEmTUI87tURElQbLfgmtXr0aWelpOGG/U+goVIU89L+FprUNoKeuLnQUonKx29cXb9PT8euvvwodhYiIPsCyX0KGhob4+eefcWrfP4iPeSl0HKoCcnNzkRjzEt3qmwodhahcvE5Jwdbr1zFr1iwe1SciqmRY9kth4cKF0NHRxqG/NgkdhaqAB7duICsnBxamLPskm9ZduQJldXUsX75c6ChERPQRlv1S0NHRwe8rVuCK2xE8C3kkdByq5K6cOAoFkQgduToJyaBHsbHYf+8efluxAtWqVRM6DhERfUQkkUgkQoeoirKzs9G0aTNo1a6DZfYuQsehSmxaj/YwRC68pk4VOgpRmRvu7IxIsRgPHz+GioqK0HGIiOgjPLJfSsrKytiwYT3uXfdC4I2rQsehSio3NxeJsa/QjSM8JIM8w8JwKSwMm7ZsYdEnIqqkWPa/wNChQ9GpU2cc2LwWubm5QsehSujetcvIFou5vj7JnByxGMs8PdHVwgJDhgwROg4REX0Cy/4XEIlE2Lx5E8IfP8AVt8NCx6FKyNv9OBRFInzLeX2SMU737iE4NhZbt22DSCQSOg4REX0Cy/4X6tixI8aPHw+nTWvwLiFe6DhUyTy+44eWdepAS1VV6ChEZSYhLQ1rvLwwfvx4tGnTRug4RET0GSz7ZWDLli1QUhBh//rfhY5ClUhubi6S3sSiW/36QkchKlPLLlyAWFERGzduFDoKEREVgWW/DNSsWRNbNm/GVfdjCLzpLXQcqiT8L19AtlgMC87rkwzxjoiAS0AANm7aBAMDA6HjEBFREVj2y8ikSZNg2a0b9vy+GJkZ6ULHoUrgmsdxKCoooL2RkdBRiMpERnY25p05gy6dOsHa2lroOEREVAws+2VEJBLh7927ER/zEq67/hQ6DlUCIff80dbQEBpckpBkxJZr1xCVmIh/7O2hoMAfH0REVQG/W5chMzMz/Prrrzi5dyeiQoOFjkMCysnJQVLcG3Tl+vokI4Jfv8YfN25g8eLFaNKkidBxiIiomFj2y9iiRYvQsGFD/P37Iq69L8f8PM8iJzcXFiz7JANyc3Mx78wZmJqaYunSpULHISKiEmDZL2OqqqrY888/CL7rj4tHDggdhwRy7dQJKCsooH3dukJHIfpiTvfuwefZM/y9Zw/U1NSEjkNERCXAsl8OLCwsYG1tDectaxEf81LoOCSAsIA7aGdkBDVlZaGjEH2R54mJWO7piUmTJqFbt25CxyEiohJi2S8nmzZtgp6uLv5aOAdisVjoOFSBsrOykJwQB0uO8FAVl5ubixnu7tCtVg3btm0TOg4REZUCy3450dfXh9O//+Khvw88HHcLHYcqkM95D87rk0zY7euLa+HhcNy/H3p6ekLHISKiUmDZL0fdu3fHwoUL4fLHBjx9cF/oOFRBbpx2h4qiItoYGgodhajUgl+/xsrLlzF37lxYWVkJHYeIiEpJJJFIJEKHkGVZWVno2KkTYhMSsfHYeahpaAgdicqZdecWaK6jDY+JE4WOQlQqmTk56Ll3L7K1tXE3IADq6upCRyIiolLikf1ypqKiAhdnZyTEvoLjuhVCx6FylpWRgeS3CZzXpypt1aVLCImLw8HDh1n0iYiqOJb9CmBmZoa//vwTnked4XvxrNBxqBzdOOsOcW4uLOrVEzoKUalcfvIE//Pxwbr169GyZUuh4xAR0Rdi2a8g1tbWGPrdd9i1fD7iY18JHYfKyc2z7lBTUkJrzutTFfQmJQXT3d3Rq2dP2NraCh2HiIjKAMt+BRGJRLDfswdaGurYsdiWV9eVUU+DAtDRxATKiopCRyEqkRyxGFOOH4dERQX79u+HggJ/PBARyQJ+N69A1atXh9O//+K+zzUc/2e70HGojGWkpSEl8S3n9alKWn35Mm5GRuLw0aOoU6eO0HGIiKiMsOxXMCsrK6xYsQKH/tyIO16eQsehMnTN4zjEEgnn9anKcX/0CH/euIENGzbwKrlERDKGZV8Av/32GwYOGoQ/F8xEdPgToeNQGfE5fwoaysr45quvhI5CVGxhcXGY6e6OEcOH4+effxY6DhERlTGusy+QpKQktO/QAalZYqw7cgoaWtpCR6IvNPlbc7SvUQOu48YJHYWoWFIyM9HTwQHQ1YXf7dvQ1ub3ISIiWcMj+wLR0dGB+8mTeBcXi78WzOIJu1VcemoKUt6947w+VRkSiQRzPDzwIiUFx0+cYNEnIpJRLPsCatSoEQ66uOC2lyeO7NgidBz6Al4njyJXIoEFyz5VEbt9fXH8wQM4ODqiSZMmQschIqJywrIvsAEDBmDNmjU4unMbL7hVhd26cAZaKipobmAgdBSiIl0MC8OvFy7gl19+wciRI4WOQ0RE5YhlvxJYsmQJho8Yge2L5yAqLEToOFQKkY8eoEu9elDk2uRUyQXFxGCyqysG9O+PDRs2CB2HiIjKGZtJJSASibDP0RH165li06wpSH6bIHQkKoGUpHdITU5CV47wUCUX/e4dRh86hEZNmsDl4EEo8uJvREQyj2W/ktDS0sLJkyeQmZqMddMnIiMtTehIVExXjh9GrkSCLlxfnyqx5MxMjDl0CIqamjh15gy0tLSEjkRERBWAZb8SadCgAc6eOYPnYcHY+vM05GRnCx2JisHP8yx0VFXRrHZtoaMQFSpHLMZkV1dEpqTgzLlz+IrXgiAikhss+5VMu3bt4Hb8OAKve2HX8gXgZRAqv+chj2FhagoFzutTJSSRSLDgzBl4RUTA9dgxNGvWTOhIRERUgdhOKqHevXtj//798DpxBAe22gkdhz4jOfEtUlKSOa9Pldb2mzfheOcOdu/ejV69egkdh4iIKpiS0AGocGPHjsXr168xb9486FWviUGTpgodiQpx6dhBSCQSWHBenyohp7t38dvFi1i6dCmsra2FjkNERAJg2a/EbG1tERsbi/Xrf4du9RroOmiY0JHoI/6XzkNPXR1NatUSOgpRPq5BQZjj4QEbGxusWbNG6DhERCQQlv1Kzs7ODjExMdixxBY6+tXQsks3oSPRB56HBqOnqSlEIpHQUYikTj1+DBs3N4wfNw47d+7k308iIjnGmf1KTiQSYc+ePejbpy82z/kJoYF3hY5E/3n75jXSOK9PlcylJ08w5dgxDBs2DHsdHHjiOBGRnONPgSpASUkJR44cRuvWrbDaegyC7/oLHYkAXHI9CAnA9fWp0rjx7BnGHTmCXr1744CzM5SU+MtbIiJ5x7JfRWhoaODc2bNo07o11vw0Fo9u+wodSe7d8bqI6hoaaFSjhtBRiHD7xQuMPngQnbp0geuxY1BRURE6EhERVQIs+1WIlpYWzp45g287dMDan35A0K0bQkeSa9FPQmBZvz7noUlwAS9fYoSLC75p3Ron3d2hpqYmdCQiIqokWParGE1NTZzy8ECXzp2xbtoEBN70FjqSXEqIjUFaaiq6coSHBHYrKgqDnZzQyNwcp8+ehaamptCRiIioEmHZr4I0NDTg4eGO7t26Yf30ibh37YrQkeTOxaPOkACw4Mm5JKArT59i2IEDaNWuHTwvX4aurq7QkYiIqJJh2a+i1NTUcOKEG3r36o0NMyfj9pWLQkeSK3e8PFFLUxP1q1UTOgrJqVOPH2P0wYOw7NEDZ8+dg7a2ttCRiIioEmLZr8JUVVVx7JgrBgwYgE1zfoSv51mhI8mNV+Fh6NagAef1SRBH7t/HxKNHMWToUJw4eRLq6upCRyIiokqKZb+KU1FRwZHDhzF06FBssbWB10lXoSPJvNfRz5GWlsYlN0kQjrdvv79g1oQJOHjoEFfdISKiz2LZlwHKyso46OKC8ePGYfuiOXDbswMSiUToWDLL86gLAPDkXKpwf924gXmnTmH27NnYu3cvFBUVhY5ERESVHK+4IiOUlJTg4OAAIyMjrF69GvExrzB56SqWgXJw1/syDLS1YaKvL3QUkhPi3FwsPX8ef/v6YtmyZVi1ahVHyIiIqFhY9mWISCTCqlWrULduXUyfPh0Jr2Mwd9MOqKpxnrcsxUY8wZBGjVi2qEKkZGbC+vhxeD55gp07d2L69OlCRyIioiqEYzwyaOrUqThx4gTuX/fC7xNHIjHujdCRZMaryAikpadzyU2qENHv3qHf/v24GR2NU6dOsegTEVGJsezLqEGDBsHb2xvvYl9i6ZiBiAoLETqSTMib17fgvD6Vs4CXL9HTwQHvFBRw4+ZN9O3bV+hIRERUBbHsy7C2bdvCz9cXNfX0sGzsYATeuCp0pCrv3rUrqKurCyM9PaGjkAw7ExyM/vv2oW7DhvD190fz5s2FjkRERFUUy76MMzY2xo0b12HRuQvWTh2H0//ac6WeL/A6Mhzd6tcXOgbJKIlEgp0+Pvjh8GH0GzgQXt7eMDAwEDoWERFVYSz7ckBHRwceHu6YPXs2HOx+w7ZfpiM9NVXoWFXO86dhSM/I4AgPlYu0rCxMO3ECS8+fx4IFC3Dk6FFoaGgIHYuIiKo4ln05oaSkhG3btuHw4cMI8L6MJaP648XTMKFjVSmX8ub1eXIulbEncXHo6eAAj9BQODs7Y8OGDVBQ4LdnIiL6cvxpImdGjRoFfz8/aKooYfHI/rhx1l3oSFVG4A0vmOjro46OjtBRSIa4P3qE7vb2EGtrw8/fH2PHjhU6EhERyRCWfTnUpEkT+Pv5YfDgQdg6bxoc7H5DdlaW0LEqvddRz9Cd8/pURrLFYiw7fx4TjhxB34ED4X/nDpo2bSp0LCIikjEs+3JKS0sLB11c8Ndff+G8yz78Pmkk4mNfCR2r0noW/AgZmZmc16cy8SopCYP//Re7/f3fj9cdOQJtbW2hYxERkQxi2ZdjIpEIs2fPhre3N5JiX2LhsD4IunVd6FiV0iXX9/P6XVj26QtdDQ+H5Z49iMzMhJeXF2xtbXk1ZiIiKjcs+4SOHTvi3t27aPVNC6ycPBr7N65CVmaG0LEqlcCb3qhfrRpq8+grlVJ6djaWnDuHIf/+i+bt2uFeYCA6d+4sdCwiIpJxLPsEAKhVqxYuXriA9evX49wBBywa0Q/hD+8LHavSiH8RxXl9KrXAV6/Q3d4eDnfvYtu2bbhw8SJq1aoldCwiIpIDLPskpaioiIULF+L27dvQ11DH4tEDcXTnNohzcoSOJqinD+8jIyuLS25SieWIxdji7Q0re3uoGxjgzt27sLW15bKaRERUYfgThwpo3rw5/Px8sWTxYhz931b8+v1guV6T/5LrQQCc16eSiUhIQP/9+7HWywsLFy3CLT8/rrZDREQVjmWfCqWiooLVq1fj5s2bEGWmY8GwPji1fw9yc3OFjlbhgnyuoVGNGqihqSl0FKoCJBIJ9t+5gy5//404BQVcu3YNa9euhYqKitDRiIhIDrHs02e1b98eAQH3MH2aDRzXrcDvk0bi9YvnQseqUPHRL9CN8/pUDOHx8Rji5IS5Hh74fvx4BAYFoVOnTkLHIiIiOcayT0VSV1fHH3/8gcuXLyM59iV+HtwDbnt2yMWFuELu3UZmNuf16fOyxWJsvXYNnXbvxnOxGOfPn8eePXugpaUldDQiIpJzLPtUbN27d0fQ/fuYZjMVB//YgPlDeyLwprfQscrVpWOHIALQxcRE6ChUSflGRaHbnj1Y6+WF2ba2CHr4EL179xY6FhEREQCWfSohHR0dbN26Fffu3UM9w6+wasoYbJlng/iYl0JHKxcPfG+ica1a0NfQEDoKVTLxqamYdfIk+jg4QL1OHfj7+2Pjxo3Q5LkdRERUibDsU6k0b94c3levwsnJCU/u+mFuf0ucsP+fzI32JMZEc16f8hHn5uLfO3fQdudOnAoPx65du3DLzw+tWrUSOhoREVEBLPtUaiKRCOPGjUNoSAhspv4El23rMX9oL9z3uSZ0tDLx0P8WMrOzYcElN+k/V54+heWePZjj4YFBw4cjOCQE06ZNg6KiotDRiIiICsWyT19MV1cX27Ztw927d2FSpzZWTh6NLfNsEPM8UuhoX+TK8ffz+p04ry/3HsbGYoSzM75zcoJuvXrw8fHB/v37Ubt2baGjERERfRbLPpWZFi1a4Jq3N5ycnBAecBtz+lngn5VLkBAbI3S0UnnofwvNDAygp64udBQSyKukJMw+eRIWf/+NZ2Ixjh07hms3buDbb78VOhoREVGxsOxTmcob7Xn65AnW2dnB77wHZvbuhH83rUby2wSh4xVbbm4uEmNecl5fTqVkZmLdlStos2MHTj97hm3btuHh48cYNmwYRCKR0PGIiIiKTSSRSCRChyDZ9e7dO2zZsgVbtm6FSEEBgyZPw8AJP0G9kq8/HnDjKlZbf48jY8eid6NGQsehCpKWlYX9d+/ij5s3kZiRgbm2tliyZAn09PSEjkZERFQqLPtUIV6/fo3169dj586dUNPUwndTZ6PP9xOgoqomdLRC/bFgFm6ccsOzRYugo1Y5M1LZScnMhOPt2/jr1i0kpKXhhx9+wMqVK1GPJ2cTEVEVx7JPFer58+dYtWoVHB0doV+zNr6zmY3u342Cqlrlmouf1qM9DJELr6lThY5C5SgpIwP2/v74n68v3mVkYOLEiVi6dCnqc3yLiIhkBMs+CSI0NBS/rViBo0eOQFtPH32+n4i+YydBt3oNoaMhNzcXY5ubYOa33+L3Xr2EjkPlIDE9HX/7+mK3nx9Ss7Nh/eOPWLRoEUy48hIREckYln0SVHh4OP744w/Y790LsVgMyyEjMWjSVBjWbyhYpjtXPWFnMwHHxo2DVUPhclDZe56YiL3+/nC4exdZubmYamODhQsXwtDQUOhoRERE5YJlnyqFhIQE7N69G39t347YmBi0694LgyZPg3m7byt89ZOtP0/HrbPuiFy8GFqqqhX62lT2JBIJbkRG4m8/P5wODoa2lhZ+mjoVv/zyCwwMDISOR0REVK5Y9qlSyczMhIuLCzZv2YJHDx+iYbNvMHjKNHzbewAUlZQqJMPUbm1gqqQIzx9/rJDXo/KRnp2No0FB+MffHw9evUITMzPMsbXFuHHjoFXJV4MiIiIqKyz7VClJJBKcP38emzZvxuVLl1CtVm1YDhmJHsNGo45pg3J73bx5/bmdOmGZlVW5vQ6Vn+eJiXC4fRv77t1DYloaBg4YgDlz58LKyopr5BMRkdxh2adKLzAwEPb29jjg7IzEt29h3qY9ug8bg459B0FdU7NMX8v34llsnG2NE+PHo1uD8tupoLKVlpWF08HBOHj/PrzCw6GtpYUp1taYOXMmGvDrSEREcoxln6qMjIwMnDhxAg4OjvD0vAg1dQ107DcIVsPHwKxVuzI5art57lT4XTyD54sXQ0NFpQxSU3mRSCTwiYrCwYAAnAwORlJ6Orp06oSJkydjzJgxHNUhIiICyz5VUZGRkdi/fz8cHBwRGfkMhqYN0H3YaHQdNAzVDeqUers/dW2Fr1VVcN7augzTUll69vYtDgUG4lBQEJ7Fx6OesTEmTJqECRMm8Cg+ERHRR1j2qUrLzc2Fl5cXHBwccOzYMWRkZKBRi1Zo17MvOvTsV6IlPHNycvBDi3r42cICS7t3L8fUVFLPEhJwKjgYp0JCcCsyElqamhgxciQmTZoECwsLKCgoCB2RiIioUmLZJ5mRmJiIU6dO4bibG86dO4f0tDQYNfga7Xv1R4ee/VC/afPPjvrcPOuOLfOmwWPiRFiYmlZgcvqYRCJBUEwMTgUH40xoKB68egVVFRX07t0bo0aPxnfffQfNMj5fg4iISBax7JNMSktLw8WLF+Hm5gZ3Dw+8TUhArTp10daqDzr07IcmbdoXWMpzw8wpuHflAp4vWQI1ZWWBkssvcW4ubkVFSQt+ZEICdHV0MHDQIHz33Xfo06cP5/CJiIhKiGWfZF52djauXbsGNzc3HHdzw8voaOjo6aNph85o0dECzTt2gYFxPfxk0RLmWpo4PWmS0JHlgkQiQfCbN/COiMDViAjciIzEu/R01DEwwNBhwzB06FB069YNytzxIiIiKjWWfZIrEokEt2/fhoeHBzw9L8HPzxdisRi16tRFfMxL9Dczg12fPjDS0xM6qkyKfPsW3hER72+RkYhNSoKKsjI6duwIq5490bt3b7Rr144z+ERERGWEZZ/kWlJSEq5evQpPT08cO3IU0TGvAAAm1aqho5EROhob45uvvoJ5rVpQqaAr+MqKzJwcPIiJwd2XL3E3Oho+L17gWXw8FBQU0KZVK/To2RNWVlbo3LkzNDQ0hI5LREQkk1j2iT4QFxeHa9euwdvbG1evXEFgUBByc3OhoqSEpgYGaGVggJZffYVv6tSBea1aUFZUFDpypZAjFiMkLg53o6Nx7+VL3I2JwcNXr5AtFkNZWRktmjVDx86dYWVlBUtLS+jr6wsdmYiISC6w7BN9RmpqKgICAnDnzp33N39/PA4JQW5uLlSVldHUwAAtatVCw+rVpTcTfX2Z3QnIzc1FdFISQt68QfCbN+//Gx+PhzExSMvKgkgkQhMzM7Tr0AHt2rVDu3bt8M0330BVVVXo6ERERHKJZZ+ohD7eAQi8dw9hT54gLT0dAKCooIB61aujob4+GlSrhobVq6NB9eow0NZGbS0t6KqplcnVfsuDRCLB2/R0vExKwqvkZLxKSkLE27d4Gh+Pp4mJCI+PR3pWFgBAQ10dTRo3hnmzZmjRogXatWuH1q1bQ1tbW+B3QURERHlY9onKgEQiwcuXLxEaGorQ0FCEhIQgNCQEocHBCI+MhFgslj5WTVkZBjo6qK2lhdoaGqj9305A3o6AlooKNP+75f2/lqoq1JSUSrSTkJWTg5SsLCRlZiIlMxPJ/91SsrKQnJmJpIyM94U+ORkxqanScp+ZnS3dhkgkgpGhIRqZmeHrRo3Q6L+bubk5jI2NeSItERFRJceyT1TOsrOzERkZiVevXhW8vXyJV9HReBUTg/i3bz+7HQWRCFpqalBSVETeP1uJRPL+9t//A4AEQLZYnK+0f0wkEkFbSwtf1a4NQyMjGNatizp16sDQ0FD6X0NDQxgYGHDpSyIioiqMZZ+oksjOzkZycjJSUlI+eUtOTkZOTg6A94U97/bxx0pKStDR0YG2tjZ0dHSkt7yPNTU1eVSeiIhIDrDsExERERHJKB7aIyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKNY9omIiIiIZBTLPhERERGRjGLZJyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKNY9omIiIiIZBTLPhERERGRjGLZJyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKNY9omIiIiIZBTLPhERERGRjGLZJyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKNY9omIiIiIZBTLPhERERGRjGLZJyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKNY9omIiIiIZBTLPhERERGRjGLZJyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKNY9omIiIiIZBTLPhERERGRjGLZJyIiIiKSUSz7REREREQyimWfiIiIiEhGsewTEREREckoln0iIiIiIhnFsk9EREREJKP+D5URSY3hFNYgAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---------------------------- mp summary statistics ----------------------------\n",
      "Age Category         Count      Percentage  \n",
      "--------------------------------------------------\n",
      "age ≤ 19             100        2.31%\n",
      "19 < age ≤ 40        1291       29.85%\n",
      "40 < age ≤ 60        2215       51.21%\n",
      "age unavailable      719        16.62%\n",
      "--------------------------------------------------------------------------------\n",
      "mp diagnoses over 60 (excluded): 20\n",
      "total women with mp: 4325\n",
      "average diagnosis age: 41.14\n",
      "% of women with mp: 1.58%\n"
     ]
    }
   ],
   "source": [
    "plot_counts_pie_chart(mp_counts)\n",
    "print_summary_statistics(mp_counts)\n",
    "print(f\"% of women with mp: {((mp_counts['n'] / 273157) * 100):.2f}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bc308ce-e411-4bf3-9f5c-f85f58fa2339",
   "metadata": {},
   "source": [
    "## MP phenotype with control group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "e08a20ca-5bde-48f4-8c11-fe0d4ba19a12",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "258842"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mp_final_df = control_df.join(mp_phenotype_df_, on=\"eid\", how=\"outer\")\n",
    "mp_final_df = mp_final_df.withColumn(\"mp_ever\", F.when(F.col(\"mp_ever\").isNull(), F.lit(0)).otherwise(F.col(\"mp_ever\")))\n",
    "mp_final_df = mp_final_df.withColumnRenamed(\"diagnosis_age\", \"mp_diagnosis_age\")\n",
    "mp_final_df = mp_final_df.select(*[\"eid\", \"mp_ever\", \"mp_diagnosis_age\", \"mp_incident_ado\", \"mp_incident_mid2\", \"mp_incident_meno2\", \"mp_over60\"])\n",
    "mp_final_df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a64519be-1842-4ed2-8cb4-3499ec7dd76e",
   "metadata": {},
   "source": [
    "# Final HMB and MP phenotype "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "3387decd-6a39-4d34-90b2-8d56b17770d1",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "258949"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_phenotype_df = hmb_final_df.join(mp_final_df, on=\"eid\", how=\"outer\")\n",
    "final_phenotype_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "7f2ae28a-9818-421f-a26e-94f38f82f3df",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "502131"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "query = f\"SELECT DISTINCT eid FROM `{ukb.database}`.`participant_0001`\"\n",
    "whole_cohort = ukb.spark.sql(query)\n",
    "whole_cohort.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ed1ade75-b46f-45be-a072-09969bc96e9d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "502131"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# join the entire cohort (inc. males)\n",
    "final_phenotype_df = final_phenotype_df.join(whole_cohort, on=\"eid\", how=\"outer\")\n",
    "final_phenotype_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8df01be-e080-4e59-b8b2-aab3bd37f5c5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pyspark.sql.functions import col\n",
    "\n",
    "# redundancy check: filter out withdrawn participants\n",
    "with open(\"withdrawn_participants.csv\", \"r\") as f:\n",
    "    withdrawn_eids = set(f.read().splitlines())\n",
    "\n",
    "final_phenotype_df = final_phenotype_df.filter(~col(\"eid\").isin(withdrawn_eids))\n",
    "final_phenotype_df.show()\n",
    "final_phenotype_df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ac6a022-af68-4891-8f21-24fc75a8ac4e",
   "metadata": {},
   "source": [
    "Final conversion to correct file format for GWAS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5e69dd9-affe-425f-9037-8a93fbd5028c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# final conversion to pandas dataframe\n",
    "hmb_gwas = final_phenotype_df.select(\"eid\", \"hmb_ever\")\n",
    "hmb_pdf = hmb_gwas.toPandas()\n",
    "hmb_pdf.columns = [\"FID\", \"hmb_ever\"]\n",
    "hmb_pdf[\"IID\"] = hmb_pdf[\"FID\"]\n",
    "\n",
    "# convert hmb_ever to 0, 1, or NA\n",
    "hmb_pdf[\"hmb_ever\"] = hmb_pdf[\"hmb_ever\"].apply(\n",
    "    lambda x: \"NA\" if pd.isnull(x) else int(x)\n",
    ")\n",
    "\n",
    "hmb_pdf = hmb_pdf[[\"FID\", \"IID\", \"hmb_ever\"]]\n",
    "print(hmb_pdf.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "ce308af7-f586-40ae-a471-3c344074eef5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hmb_phenotype.tsv saved successfully\n"
     ]
    }
   ],
   "source": [
    "def write_df_to_file(pdf, filename):\n",
    "    pdf.to_csv(filename, sep=\"\\t\", index=False, encoding=\"utf-8\")\n",
    "    print(f\"{filename} saved successfully\")\n",
    "    \n",
    "write_df_to_file(hmb_pdf, \"hmb_phenotype.tsv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "940e5b68-e33e-45cb-99a0-b418345c96c2",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# final conversion to pandas dataframe\n",
    "mp_gwas = final_phenotype_df.select(\"eid\", \"mp_ever\")\n",
    "mp_pdf = mp_gwas.toPandas()\n",
    "mp_pdf.columns = [\"FID\", \"mp_ever\"]\n",
    "mp_pdf[\"IID\"] = mp_pdf[\"FID\"]\n",
    "\n",
    "# convert mp_ever to 0, 1, or NA\n",
    "mp_pdf[\"mp_ever\"] = mp_pdf[\"mp_ever\"].apply(\n",
    "    lambda x: \"NA\" if pd.isnull(x) else int(x)\n",
    ")\n",
    "\n",
    "mp_pdf = mp_pdf[[\"FID\", \"IID\", \"mp_ever\"]]\n",
    "print(mp_pdf.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "4cc3f7cb-e398-4db8-955a-113289f8564b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mp_phenotype.tsv saved successfully\n"
     ]
    }
   ],
   "source": [
    "write_df_to_file(mp_pdf, \"mp_phenotype.tsv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca5180c4-9393-4184-a860-fdc5cd8352eb",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "d7512726-ce9b-453d-a31d-efd82705974c",
   "metadata": {},
   "source": [
    "## Removing existing conditions for sensitivity analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afebaa1f-84e1-425e-8ab9-8ecdb960def2",
   "metadata": {},
   "source": [
    "Collecting participants with exclusion criteria for sensitivity analysis:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "fcd2b607-6b29-464f-92f7-1f18783b9cf0",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "filter_females = lambda df: females_df.join(df, on=\"eid\", how=\"inner\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "d514de16-7ba9-45c2-87ce-82c732e79553",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 9 ICD10 codes...\n",
      "Found 3353 matching hospital records\n",
      "Found 1 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 3353 matching records in 26.31 seconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "3353"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd10s = [f\"N80.{i}\" for i in range(1,10)] # Endometriosis N80.1 – N80.9\n",
    "endometriosis_df = ukb.get_icd10(icd10s)\n",
    "endometriosis_df = filter_females(endometriosis_df).select(\"eid\")\n",
    "endometriosis_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "a9c79221-bfb0-48c1-8ac3-7a763b67d08f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 1 ICD10 codes...\n",
      "Found 444 matching hospital records\n",
      "Found 0 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 444 matching records in 26.91 seconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "442"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd10s = [\"E28.2\"] # Polycystic Ovary Syndrome E28.2\n",
    "POS_df = ukb.get_icd10(icd10s)\n",
    "POS_df = filter_females(POS_df).select(\"eid\")\n",
    "POS_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "c7aff674-7f67-4cfd-b2cc-ecb5760cce19",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 10 ICD10 codes...\n",
      "Found 15764 matching hospital records\n",
      "Found 1 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 15764 matching records in 24.33 seconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "15763"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd10s = [f\"D25.{i}\" for i in range(10)]  # Uterine Leiomyoma D25\n",
    "uterine_leiomyoma_df = ukb.get_icd10(icd10s)\n",
    "uterine_leiomyoma_df = filter_females(uterine_leiomyoma_df).select(\"eid\")\n",
    "uterine_leiomyoma_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "516f6050-b467-496a-9292-6c39a71ee44c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 1 ICD10 codes...\n",
      "Found 12171 matching hospital records\n",
      "Found 0 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 12171 matching records in 26.01 seconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "12171"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd10s = [\"N84.0\"] # Uterine Polyps N84.0\n",
    "uterine_polyps_df = ukb.get_icd10(icd10s)\n",
    "uterine_polyps_df = filter_females(uterine_polyps_df).select(\"eid\")\n",
    "uterine_polyps_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "0d0f72e3-4ac2-49c3-a33d-5abecd1add92",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 1 ICD10 codes...\n",
      "Found 2692 matching hospital records\n",
      "Found 0 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 2692 matching records in 25.44 seconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "2692"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd10s = [\"N80.0\"] # Adenomyosis N80.0\n",
    "adenomyosis_df = ukb.get_icd10(icd10s)\n",
    "adenomyosis_df = filter_females(adenomyosis_df).select(\"eid\")\n",
    "adenomyosis_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "0b514391-0373-4a5e-9c7d-30cbed87ed04",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Searching 80 ICD10 codes...\n",
      "Found 8113 matching hospital records\n",
      "Found 9 matching death records\n",
      "Found 0 matching cancer records\n",
      "Operation complete: Found 8118 matching records in 26.01 seconds.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "8114"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "icd10s = [f\"N7{i}.{a}\" for i in range(8) for a in range(10)] # Pelvic Inflammatory Disease N70 – N77\n",
    "PID_df = ukb.get_icd10(icd10s)\n",
    "PID_df = filter_females(PID_df).select(\"eid\")\n",
    "PID_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "dc5104c2-ba33-48ef-b47e-8b03f59d1265",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "excluded_df = endometriosis_df.join(POS_df, on=\"eid\", how=\"outer\")\n",
    "excluded_df = excluded_df.join(uterine_leiomyoma_df, on=\"eid\", how=\"outer\")\n",
    "excluded_df = excluded_df.join(uterine_polyps_df, on=\"eid\", how=\"outer\")\n",
    "excluded_df = excluded_df.join(adenomyosis_df, on=\"eid\", how=\"outer\")\n",
    "excluded_df = excluded_df.join(PID_df, on=\"eid\", how=\"outer\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "eab47b2b-e80b-4d5e-8fd7-ab721606aeda",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33739"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "excluded_df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25216633-2c13-47dc-93f9-77ebd658e155",
   "metadata": {},
   "source": [
    "Excluding participants and deriving variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "4c0859ab-7c1f-4cfd-8365-e90b2e694965",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20218"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "excluded_hmb_phenotype_df = hmb_phenotype_df.join(excluded_df, on=\"eid\", how=\"anti\")\n",
    "excluded_hmb_phenotype_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "8149566e-7404-4400-b558-a4884cbd26ea",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2861"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "excluded_mp_phenotype_df = mp_phenotype_df.join(excluded_df, on=\"eid\", how=\"anti\")\n",
    "excluded_mp_phenotype_df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "a1785255-e309-4f85-9e92-aaeda8df6eac",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---------------------------- hmb_excluded summary statistics ----------------------------\n",
      "Age Category         Count      Percentage  \n",
      "--------------------------------------------------\n",
      "age ≤ 19             60         0.30%\n",
      "19 < age ≤ 40        4845       24.04%\n",
      "40 < age ≤ 60        14969      74.26%\n",
      "age unavailable      283        1.40%\n",
      "--------------------------------------------------------------------------------\n",
      "hmb_excluded diagnoses over 60 (excluded): 61\n",
      "total women with hmb_excluded: 20157\n",
      "average diagnosis age: 44.22\n"
     ]
    }
   ],
   "source": [
    "excluded_hmb_phenotype_df_, excluded_hmb_counts = derive_variables(excluded_hmb_phenotype_df, \"hmb_excluded\")\n",
    "print_summary_statistics(excluded_hmb_counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "523fac6a-33a0-4ad9-976f-42ace66261e2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "---------------------------- mp_excluded summary statistics ----------------------------\n",
      "Age Category         Count      Percentage  \n",
      "--------------------------------------------------\n",
      "age ≤ 19             87         3.06%\n",
      "19 < age ≤ 40        917        32.23%\n",
      "40 < age ≤ 60        1312       46.12%\n",
      "age unavailable      529        18.59%\n",
      "--------------------------------------------------------------------------------\n",
      "mp_excluded diagnoses over 60 (excluded): 16\n",
      "total women with mp_excluded: 2845\n",
      "average diagnosis age: 40.19\n"
     ]
    }
   ],
   "source": [
    "excluded_mp_phenotype_df_, excluded_mp_counts = derive_variables(excluded_mp_phenotype_df, \"mp_excluded\")\n",
    "print_summary_statistics(excluded_mp_counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a9ddcab3-da40-440b-9870-15a0160a381c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "727a7e69-1275-43ab-a9f6-00be05de9689",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eddfc874-1b04-4abc-8d63-42ba2ace4e1f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "635577cd-bb26-472d-ba48-bade9fcd5dcf",
   "metadata": {},
   "source": [
    "## Final summary statistics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1328780d-0c94-484f-8dc8-37c8c0815aba",
   "metadata": {},
   "source": [
    "Both <- hmb_ever==1 AND mp_ever==1\n",
    "\n",
    "HMBonly <- hmb_ever==1 AND mp_ever==0\n",
    "\n",
    "MPonly <- hmb_ever==0 AND mp_ever==1\n",
    "\n",
    "Neither <- hmb_ever==0 AND mp_ever==0\n",
    "\n",
    "If either hmb_ever or mp_ever is missing, the participant can be coded as missing for this variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "ba670a51-41bb-4118-aaf9-2e95285353c9",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "final_summary_stats = (\n",
    "    final_phenotype_df.agg(\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 1) & (F.col(\"mp_ever\") == 1), 1).otherwise(0),\n",
    "        ).alias(\"both\"),\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 1) & (F.col(\"mp_ever\") == 0), 1).otherwise(0),\n",
    "        ).alias(\"hmb_only\"),\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 0) & (F.col(\"mp_ever\") == 1), 1).otherwise(0),\n",
    "        ).alias(\"mp_only\"),\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 0) & (F.col(\"mp_ever\") == 0), 1).otherwise(0),\n",
    "        ).alias(\"neither\"),\n",
    "    )\n",
    "    .collect()[0]\n",
    "    .asDict()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "8f878503-b1c8-4d84-8d03-77ba7bad2bc1",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "both: 2578\n",
      "hmb_only: 26971\n",
      "mp_only: 1706\n",
      "neither: 227546\n"
     ]
    }
   ],
   "source": [
    "for key, value in final_summary_stats.items():\n",
    "    print(f\"{key}: {value}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fc719d7-b43e-4b3a-ab86-2bb6ed111805",
   "metadata": {},
   "source": [
    "Final summary statistics for participants with known disorders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "a9c95718-0f64-4bf4-93f9-82a83a6536a6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "known_disorder_df = final_phenotype_df.join(excluded_df, on=\"eid\", how=\"inner\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "56c0e80a-2b50-450f-989b-918e50782b40",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "known_disorder_final_summary_stats = (\n",
    "    known_disorder_df.agg(\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 1) & (F.col(\"mp_ever\") == 1), 1).otherwise(0),\n",
    "        ).alias(\"both\"),\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 1) & (F.col(\"mp_ever\") == 0), 1).otherwise(0),\n",
    "        ).alias(\"hmb_only\"),\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 0) & (F.col(\"mp_ever\") == 1), 1).otherwise(0),\n",
    "        ).alias(\"mp_only\"),\n",
    "        F.sum(\n",
    "            F.when((F.col(\"hmb_ever\") == 0) & (F.col(\"mp_ever\") == 0), 1).otherwise(0),\n",
    "        ).alias(\"neither\"),\n",
    "    )\n",
    "    .collect()[0]\n",
    "    .asDict()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "e29be5a6-ca77-42b8-aa7d-385c15bda266",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "both: 1094\n",
      "hmb_only: 8404\n",
      "mp_only: 386\n",
      "neither: 23855\n"
     ]
    }
   ],
   "source": [
    "for key, value in known_disorder_final_summary_stats.items():\n",
    "    print(f\"{key}: {value}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "483a30f1",
   "metadata": {},
   "source": [
    "# Age distribution curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "2790b4c4",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "\n",
    "# filter out rows where 'diagnosis_age' is null\n",
    "filtered_ages = (\n",
    "    mp_phenotype_df.filter(mp_phenotype_df[\"diagnosis_age\"].isNotNull())\n",
    "    .select(\"diagnosis_age\")\n",
    "    .rdd.flatMap(lambda x: x)\n",
    "    .collect()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "d0948dd4",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_theme(style=\"whitegrid\")  \n",
    "plt.figure(figsize=(10, 6))  \n",
    "\n",
    "# calculate average diagnosis age and plot age distribution curve\n",
    "avg_diagnosis_age = mp_counts[\"avg_diagnosis_age\"]\n",
    "\n",
    "sns.kdeplot(filtered_ages, fill=True, clip=(15, 70), color='skyblue', linewidth=2)\n",
    "plt.axvline(avg_diagnosis_age, color='red', linestyle='--', linewidth=2, label=f\"Mean = {avg_diagnosis_age:.2f}\")\n",
    "\n",
    "plt.title(\"Menstrual Pain Age Distribution Curve\", fontsize=16)\n",
    "plt.xlabel(\"Age\", fontsize=12)\n",
    "plt.ylabel(\"Density\", fontsize=12)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c5853332",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "de02e1ea",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "51921140",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7602bc21",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a88e1b32",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}