1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
|
#!/usr/bin/env python3
from Bio import Entrez
Entrez.email = 'another_email@gmail.com'
import xml.etree.ElementTree as ET
import json
import os
from dateutil import parser
num_ids_for_request = 100
dir_metadata = 'metadata_from_nuccore'
dir_fasta_and_yaml = 'fasta_and_yaml'
dir_dict_ontology_standardization = 'dict_ontology_standardization/'
path_ncbi_virus_accession = 'sequences.acc'
def chunks(lst, n):
for i in range(0, len(lst), n):
yield lst[i:i + n]
if not os.path.exists(dir_metadata):
os.makedirs(dir_metadata)
# Take all the ids
id_set = set()
term_list = ['SARS-CoV-2', 'SARS-CoV2', 'SARS CoV2', 'SARSCoV2', 'txid2697049[Organism]']
for term in term_list:
tmp_list = Entrez.read(
Entrez.esearch(db='nuccore', term=term, idtype='acc', retmax='10000')
)['IdList']
# Remove mRNAs, ncRNAs, Proteins, and predicted models (more information here: https://en.wikipedia.org/wiki/RefSeq)
tmp_list = [x for x in tmp_list if x[:2] not in ['NM', 'NR', 'NP', 'XM', 'XR', 'XP', 'WP']]
# Remove the version in the id
tmp_list = [x.split('.')[0] for x in tmp_list]
print(term, len(tmp_list))
#tmp_list = tmp_list[0:2] # restricting to small run
id_set.update([x.split('.')[0] for x in tmp_list])
print(term_list, len(id_set))
if os.path.exists(path_ncbi_virus_accession):
with open(path_ncbi_virus_accession) as f:
tmp_list = [line.strip('\n') for line in f]
print('NCBI Virus', len(tmp_list))
id_set.update(tmp_list)
term_list.append('NCBI Virus')
print(term_list, len(id_set))
for i, id_x_list in enumerate(chunks(list(id_set), num_ids_for_request)):
path_metadata_xxx_xml = os.path.join(dir_metadata, 'metadata_{}.xml'.format(i))
print('Requesting {} ids --> {}'.format(len(id_x_list), path_metadata_xxx_xml))
with open(path_metadata_xxx_xml, 'w') as fw:
fw.write(
Entrez.efetch(db='nuccore', id=id_x_list, retmode='xml').read()
)
term_to_uri_dict = {}
for path_dict_xxx_csv in [os.path.join(dir_dict_ontology_standardization, name_xxx_csv) for name_xxx_csv in os.listdir(dir_dict_ontology_standardization) if name_xxx_csv.endswith('.csv')]:
print('Read {}'.format(path_dict_xxx_csv))
with open(path_dict_xxx_csv) as f:
for line in f:
if len(line.split(',')) > 2:
term, uri = line.strip('\n').split('",')
term = term.strip('"')
else:
term, uri = line.strip('\n').split(',')
term_to_uri_dict[term] = uri
species_to_taxid_dict = {
'Homo sapiens': 'http://purl.obolibrary.org/obo/NCBITaxon_9606'
}
if not os.path.exists(dir_fasta_and_yaml):
os.makedirs(dir_fasta_and_yaml)
missing_value_list = []
for path_metadata_xxx_xml in [os.path.join(dir_metadata, name_metadata_xxx_xml) for name_metadata_xxx_xml in os.listdir(dir_metadata) if name_metadata_xxx_xml.endswith('.xml')]:
tree = ET.parse(path_metadata_xxx_xml)
GBSet = tree.getroot()
for GBSeq in GBSet:
accession_version = GBSeq.find('GBSeq_accession-version').text
GBSeq_sequence = GBSeq.find('GBSeq_sequence')
if GBSeq_sequence is None:
print(accession_version, ' - sequence not found')
continue
# A general default-empty yaml could be read from the definitive one
info_for_yaml_dict = {
'id': 'placeholder',
'host': {},
'sample': {},
'virus': {},
'technology': {},
'submitter': {}
}
info_for_yaml_dict['sample']['sample_id'] = accession_version
info_for_yaml_dict['sample']['source_database_accession'] = ["http://identifiers.org/insdc/"+accession_version+"#sequence"] #accession is turned into resolvable URL/URI now
# submitter info
GBSeq_references = GBSeq.find('GBSeq_references')
if GBSeq_references is not None:
info_for_yaml_dict['submitter']['authors'] = ["{}".format(x.text) for x in GBSeq_references.iter('GBAuthor')]
GBReference = GBSeq_references.find('GBReference')
if GBReference is not None:
GBReference_journal = GBReference.find('GBReference_journal')
if GBReference_journal is not None and GBReference_journal.text != 'Unpublished':
if 'Submitted' in GBReference_journal.text:
info_for_yaml_dict['submitter']['submitter_name'] = ["{}".format(GBReference_journal.text.split(') ')[1].split(',')[0].strip())]
info_for_yaml_dict['submitter']['submitter_address'] = ','.join(GBReference_journal.text.split(') ')[1].split(',')[1:]).strip()
else:
info_for_yaml_dict['submitter']['additional_submitter_information'] = GBReference_journal.text
GBSeq_comment = GBSeq.find('GBSeq_comment')
if GBSeq_comment is not None and 'Assembly-Data' in GBSeq_comment.text:
GBSeq_comment_text = GBSeq_comment.text.split('##Assembly-Data-START## ; ')[1].split(' ; ##Assembly-Data-END##')[0]
for info_to_check, field_in_yaml in zip(
['Assembly Method', 'Coverage', 'Sequencing Technology'],
['sequence_assembly_method', 'sequencing_coverage', 'sample_sequencing_technology']
):
if info_to_check in GBSeq_comment_text:
tech_info_to_parse = GBSeq_comment_text.split('{} :: '.format(info_to_check))[1].split(' ;')[0]
if field_in_yaml == 'sequencing_coverage':
# A regular expression would be better!
try:
info_for_yaml_dict['technology'][field_in_yaml] = [
float(tech_info_to_parse.strip('(average)').strip("reads/nt").strip('(average for 6 sequences)').replace(',', '.').strip(' xX>'))
]
except ValueError:
print(accession_version, "Couldn't make sense of Coverage '%s'" % tech_info_to_parse)
pass
elif field_in_yaml == 'sample_sequencing_technology':
new_seq_tec_list = []
for seq_tec in tech_info_to_parse.split(';'):
seq_tec = seq_tec.strip()
if seq_tec in term_to_uri_dict:
seq_tec = term_to_uri_dict[seq_tec]
else:
missing_value_list.append('\t'.join([accession_version, 'sample_sequencing_technology', seq_tec]))
new_seq_tec_list.append(seq_tec)
info_for_yaml_dict['technology']['sample_sequencing_technology'] = [x for x in new_seq_tec_list]
else:
info_for_yaml_dict['technology'][field_in_yaml] = tech_info_to_parse
for GBFeature in GBSeq.iter('GBFeature'):
if GBFeature.find('GBFeature_key').text != 'source':
continue
for GBQualifier in GBFeature.iter('GBQualifier'):
GBQualifier_value = GBQualifier.find('GBQualifier_value')
if GBQualifier_value is None:
continue
GBQualifier_value_text = GBQualifier_value.text
GBQualifier_name_text = GBQualifier.find('GBQualifier_name').text
if GBQualifier_name_text == 'host':
GBQualifier_value_text_list = GBQualifier_value_text.split('; ')
if GBQualifier_value_text_list[0] in species_to_taxid_dict:
info_for_yaml_dict['host']['host_species'] = species_to_taxid_dict[GBQualifier_value_text_list[0]]
if len(GBQualifier_value_text_list) > 1:
if GBQualifier_value_text_list[1] in ['male', 'female']:
if GBQualifier_value_text_list[1]=='male':
info_for_yaml_dict['host']['host_sex'] = "http://purl.obolibrary.org/obo/PATO_0000384"
elif GBQualifier_value_text_list[1]=='female':
info_for_yaml_dict['host']['host_sex'] = "http://purl.obolibrary.org/obo/PATO_0000383"
elif GBQualifier_value_text_list[1] in term_to_uri_dict:
info_for_yaml_dict['host']['host_health_status'] = term_to_uri_dict[GBQualifier_value_text_list[1]]
else:
missing_value_list.append('\t'.join([accession_version, GBQualifier_name_text, GBQualifier_value_text_list[1]]))
if 'age' in GBQualifier_value_text:
info_for_yaml_dict['host']['host_age'] = int(GBQualifier_value_text_list[2].split('age ')[1])
info_for_yaml_dict['host']['host_age_unit'] = 'http://purl.obolibrary.org/obo/UO_0000036'
elif GBQualifier_name_text == 'collected_by':
if any([x in GBQualifier_value_text.lower() for x in ['institute', 'hospital', 'city', 'center']]):
info_for_yaml_dict['sample']['collecting_institution'] = GBQualifier_value_text
else:
info_for_yaml_dict['sample']['collector_name'] = GBQualifier_value_text
elif GBQualifier_name_text == 'isolation_source':
if GBQualifier_value_text.upper() in term_to_uri_dict:
GBQualifier_value_text = GBQualifier_value_text.upper() # For example, in case of 'usa: wa'
if GBQualifier_value_text in term_to_uri_dict:
info_for_yaml_dict['sample']['specimen_source'] = [term_to_uri_dict[GBQualifier_value_text]]
else:
if GBQualifier_value_text in ['NP/OP swab', 'nasopharyngeal and oropharyngeal swab', 'nasopharyngeal/oropharyngeal swab', 'np/np swab', 'np/op']:
info_for_yaml_dict['sample']['specimen_source'] = [term_to_uri_dict['nasopharyngeal swab'], term_to_uri_dict['oropharyngeal swab']]
elif GBQualifier_value_text in ['nasopharyngeal swab/throat swab', 'nasopharyngeal/throat swab']:
info_for_yaml_dict['sample']['specimen_source'] = [term_to_uri_dict['nasopharyngeal swab'], term_to_uri_dict['throat swab']]
elif GBQualifier_value_text in ['nasopharyngeal aspirate/throat swab']:
info_for_yaml_dict['sample']['specimen_source'] = [term_to_uri_dict['nasopharyngeal aspirate'], term_to_uri_dict['throat swab']]
else:
missing_value_list.append('\t'.join([accession_version, 'specimen_source', GBQualifier_value_text]))
elif GBQualifier_name_text == 'collection_date':
# TO_DO: which format we will use?
date_to_write = GBQualifier_value_text
if len(GBQualifier_value_text.split('-')) == 1:
if int(GBQualifier_value_text) < 2020:
date_to_write = "15 12 {}".format(GBQualifier_value_text)
else:
date_to_write = "15 01 {}".format(GBQualifier_value_text)
if 'additional_collection_information' in info_for_yaml_dict['sample']:
info_for_yaml_dict['sample']['additional_collection_information'] += "; The 'collection_date' is estimated (the original date was: {})".format(GBQualifier_value_text)
else:
info_for_yaml_dict['sample']['additional_collection_information'] = "The 'collection_date' is estimated (the original date was: {})".format(GBQualifier_value_text)
elif len(GBQualifier_value_text.split('-')) == 2:
date_to_write += '-15'
if 'additional_collection_information' in info_for_yaml_dict['sample']:
info_for_yaml_dict['sample']['additional_collection_information'] += "; The 'collection_date' is estimated (the original date was: {})".format(GBQualifier_value_text)
else:
info_for_yaml_dict['sample']['additional_collection_information'] = "The 'collection_date' is estimated (the original date was: {})".format(GBQualifier_value_text)
elif len(GBQualifier_value_text.split('-')) == 3:
GBQualifier_value_text_list = GBQualifier_value_text.split('-')
if GBQualifier_value_text_list[1].isalpha():
date_to_write = GBQualifier_value_text_list[1] + ' ' + GBQualifier_value_text_list[0] + ' ' + GBQualifier_value_text_list[2]
info_for_yaml_dict['sample']['collection_date'] = date_to_write
elif GBQualifier_name_text in ['lat_lon', 'country']:
if GBQualifier_value_text == 'Hong Kong':
GBQualifier_value_text = 'China: Hong Kong'
if GBQualifier_value_text in term_to_uri_dict:
GBQualifier_value_text = term_to_uri_dict[GBQualifier_value_text]
else:
missing_value_list.append('\t'.join([accession_version, GBQualifier_name_text, GBQualifier_value_text]))
info_for_yaml_dict['sample']['collection_location'] = GBQualifier_value_text
elif GBQualifier_name_text == 'note':
if 'additional_collection_information' in info_for_yaml_dict['sample']:
info_for_yaml_dict['sample']['additional_collection_information'] += '; ' + GBQualifier_value_text
else:
info_for_yaml_dict['sample']['additional_collection_information'] = GBQualifier_value_text
elif GBQualifier_name_text == 'isolate':
info_for_yaml_dict['virus']['virus_strain'] = GBQualifier_value_text
elif GBQualifier_name_text == 'db_xref':
info_for_yaml_dict['virus']['virus_species'] = "http://purl.obolibrary.org/obo/NCBITaxon_"+GBQualifier_value_text.split('taxon:')[1]
# Remove technology key if empty!
if (info_for_yaml_dict['technology']=={}):
del info_for_yaml_dict['technology']
with open(os.path.join(dir_fasta_and_yaml, '{}.fasta'.format(accession_version)), 'w') as fw:
fw.write('>{}\n{}'.format(accession_version, GBSeq_sequence.text.upper()))
with open(os.path.join(dir_fasta_and_yaml, '{}.yaml'.format(accession_version)), 'w') as fw:
json.dump(info_for_yaml_dict, fw, indent=2)
if len(missing_value_list) > 0:
with open('missing_terms.tsv', 'w') as fw:
fw.write('\n'.join(missing_value_list))
|
