-
Notifications
You must be signed in to change notification settings - Fork 0
/
query_umls.py
256 lines (182 loc) · 7.51 KB
/
query_umls.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
# -*- coding: utf-8 -*-
"""
Created on Wed Apr 28 13:44:27 2021
*** Search the UMLS Metathesaurus via the REST API ***
You will first need to do the following:
- Register for an account https://www.nlm.nih.gov/databases/umls.html
- Log into your account and copy the API key from the 'My Profile' area.
- Copy/paste the API key into the code below or import it via a lib etc.
Usage:
Search the Metathesaurus for strings.
Optionally: Specify one or more UMLS vocabs to use
List of available vocabularies and abbreviations:
https://www.nlm.nih.gov/research/umls/sourcereleasedocs/index.html
UMLS search parameters:
https://documentation.uts.nlm.nih.gov/rest/search/
UMLS search endpoints:
https://documentation.uts.nlm.nih.gov/rest/home.html
"""
# Imports ####################################################################
import requests
import json
import pandas as pd
from lxml.html import fromstring
# Personal API key
import umls_api_key
# Globals ####################################################################
uri="https://utslogin.nlm.nih.gov"
auth_endpoint = "/cas/v1/api-key"
# Class defs #################################################################
class Authentication():
'''
https://documentation.uts.nlm.nih.gov/rest/authentication.html
'''
def __init__(self, apikey):
self.apikey=apikey
self.service = "http://umlsks.nlm.nih.gov"
def get_tgt(self):
'''
Request a Ticket Granting Ticket (TGT)
Valid for 8 hours
'''
params = {'apikey': self.apikey}
h = {"Content-type": "application/x-www-form-urlencoded",
"Accept": "text/plain",
"User-Agent": "python" }
r = requests.post(uri+auth_endpoint,data=params,headers=h)
response = fromstring(r.text)
tgt = response.xpath('//form/@action')[0]
return tgt
def get_st(self, tgt):
'''
Request a Service Ticket
Expires after single use, or 5 mins after generation if not used.
'''
params = {'service': self.service}
h = {"Content-type": "application/x-www-form-urlencoded",
"Accept": "text/plain",
"User-Agent":"python" }
r = requests.post(tgt,data=params,headers=h)
st = r.text
return st
class searchUMLS():
def __init__(self, apikey):
self.apikey = apikey
self.auth = Authentication(apikey)
def search_term(self, search_str, vocab=None, id_type='concept', num_results=1e6, as_df=True):
"""
Args
----------
search_str : str
Search string
vocab : str, optional
Specify vocabs as a comma sep string.
The default is None; will return results for all UMLS vocabs
id_type : str, optional
. The default is 'concept'. See UMLS search docs for more info.
num_results : int or float, optional
Number of results to return. The default is 1e6.
Returns
-------
results : list
returns a list of json dicts.
"""
n = int(num_results)
results = []
tgt = self.auth.get_tgt()
uri = "https://uts-ws.nlm.nih.gov/rest/"
content_endpoint = "search/current"
page = 0
while True:
service_ticket = self.auth.get_st(tgt) # New st needed per page
page += 1
query = {'string':search_str,
'ticket':service_ticket,
'pageNumber':page,
'sabs': vocab,
'returnIdType':id_type}
r = requests.get(uri+content_endpoint, params=query)
r.encoding = 'utf-8'
items = json.loads(r.text)
json_data = items['result']
if isinstance(json_data, dict):
if json_data["results"][0]["ui"] == "NONE":
break
else:
results.extend(json_data['results'])
res_count = len(results)
if res_count >= n:
break
else:
break
results = results[0:n]
if as_df:
results = self.get_df(results)
return results
def search_cui(self, cui_lst, as_df=True):
'''
https://www.nlm.nih.gov/research/umls/META3_current_semantic_types.html
'''
results = []
for cui in cui_lst:
print(cui)
tgt = self.auth.get_tgt()
uri = "https://uts-ws.nlm.nih.gov/rest/"
content_endpoint = "content/current"
#print
#for cui in cui_list:
service_ticket = self.auth.get_st(tgt) # New st needed per page
query = {'ticket':service_ticket}
r = requests.get(uri+content_endpoint+"/CUI/"+cui, params=query)
r.encoding = 'utf-8'
items = json.loads(r.text)
if 'error' in items:
print(f"No results found for {cui}")
continue
else:
results.append(items['result'])
if as_df:
results = self.get_df(results)
return results
def get_df(self, json_data):
'''
Converts list of json dicts to pandas df
'''
df = pd.DataFrame(json_data)
# Expand nested semantic types dict
if 'semanticTypes' in df.columns:
df[['semantic_type', 'TUI']] = df['semanticTypes'].str[0].apply(pd.Series)
df.drop('semanticTypes', inplace=True, axis=1)
else:
pass
return df
if __name__ == "__main__":
# Text search #############################################################
# Specify a search term or code
search_term = 'kidney stone'
# Load personal API key from file (or copy/paste etc.)
apikey = umls_api_key.key()
# Instantiate client
client = searchUMLS(apikey)
# Get dataframe of search results (include MeSH terms only)
df1 = client.search_term(search_term, vocab='MSH, MTH', num_results=200)
display(df1)
# Get dataframe of search results (include all terms)
df2 = client.search_term(search_term)
display(df2)
# Concept search (including semantic type) ################################
# Specify a list of concept IDs
concept_ids = ['C0009044','C2097260', 'x']
# Ger dataframe of all valid UMLS concepts
df3 = client.search_cui(concept_ids)
display(df3)
# Get CUIs from previous term search
concept_ids = list(df2['ui'].values)
df4 = client.search_cui(concept_ids)
display(df4)
print(df4.head(1).T)
# Merge the term and cui search results; dump to csv
df5 = df4.merge(df2, on='ui')
df5 = df5[['ui', 'name_x', 'semantic_type', 'rootSource']].copy()
df5.columns = ['cui', 'name', 'semantic_type', 'vocab']
df5.to_csv(f'umls_{search_term}.csv')