def fetch_uniprot_data(accessions):
"""
Fetch UniProt protein descriptions, gene names, sequences, etc.
All information is stored in UNIPROT_DATA and can be accessed with
:func:`.get_uniprot_data`.
Parameters
----------
accessions : list of str
"""
accessions = set(accessions).difference(UNIPROT_DATA)
if not accessions:
return {}
cache_dir = tempfile.mkdtemp(suffix="uniprot")
UNIPROT_DATA.update(
uniprot.get_metadata_with_some_seqid_conversions(
accessions,
cache_dir=cache_dir,
))
shutil.rmtree(cache_dir)
return {i: UNIPROT_DATA.get(i, {}) for i in accessions}
def fill_protein_sequences(proteins):
print('Filling protein sequences:')
for i in range(0, len(proteins)):
# 1. send requests while server will not get correct response
server_response = None
while not server_response:
server_response = uniprot.get_metadata_with_some_seqid_conversions([proteins[i].id])
# 2. store sequence in protein
proteins[i].sequence = server_response[proteins[i].id]['sequence']
print('Filling protein sequences: done')
def load_sequences_from_uniprot(proteins, clean_seqid=None, cache_basename=None):
if clean_seqid:
change_seqids_in_proteins(proteins, clean_seqid)
seqids = []
for seqid in proteins:
seqids.append(seqid)
if 'other_seqids' in proteins[seqid]['attr']:
seqids.extend(proteins[seqid]['attr']['other_seqids'])
uniprot_data = uniprot.get_metadata_with_some_seqid_conversions(seqids, cache_basename)
load_fastas_into_proteins(proteins, uniprot_data)
if cache_basename:
uniprot.write_fasta(cache_basename+'.fasta', uniprot_data, uniprot_data.keys())
def add_uniprot_data(protxml_groups, cache_file=None):
"""
Processes the data from an PROTXML file, reads the
seqids, and attempts to mapt to a UNIPROT ID
and then, fetch the metadata of that protein from
the uniprot.org website: organism, gene, description
etc.
"""
seqids = get_all_seqids(protxml_groups)
if is_url_connected('http://uniprot.org'):
uniprot_dict = uniprot.get_metadata_with_some_seqid_conversions(
seqids, cache_file)
else:
print "Can't connect to www.uniprot.org, won't use uniprot metatdata"
uniprot_dict = {}
for group_id, protxml_group in protxml_groups.items():
for protein in protxml_group['proteins']:
protein['id'] = ''
protein['acc'] = protein['protein_name']
names = [protein['protein_name']] + protein['other_seqids']
new_seqid = uniprot.sort_seqids_by_uniprot(names, uniprot_dict)[0]
if new_seqid != protein['protein_name']:
print "Protein group %s%s is better represented with %s than %s" % \
(group_id,
protein['group_sibling_id'],
uniprot.get_naked_seqid(new_seqid),
uniprot.get_naked_seqid(protein['protein_name']))
protein['protein_name'] = new_seqid
protein['other_seqids'] = names[1:]
protein['acc'] = new_seqid
protein['other_seqids'] = ';'.join(protein['other_seqids'])
if new_seqid not in uniprot_dict:
print "No uniprot metadata for protein group %s%s seqid %s" % \
(group_id,
protein['group_sibling_id'],
uniprot.get_naked_seqid(new_seqid))
continue
protein['link'] = ''
uniprot_entry = uniprot_dict[new_seqid]
protein['id'] = uniprot_entry['id']
protein['acc'] = uniprot_entry['accs'][0]
protein['link'] = \
'=HYPERLINK("http://uniprot.org/uniprot/%s")' % \
uniprot_dict[new_seqid]['id']
if 'gene' in uniprot_entry:
protein['gene'] = uniprot_entry['gene']
if 'organism' in uniprot_entry:
protein['organism'] = uniprot_entry['organism']
protein['description'] = '; '.join(uniprot_entry['descriptions'])
if 'length' in uniprot_entry:
protein['length'] = uniprot_entry['length']
def add_uniprot_data(protxml_groups, cache_file=None):
"""
Processes the data from an PROTXML file, reads the
seqids, and attempts to mapt to a UNIPROT ID
and then, fetch the metadata of that protein from
the uniprot.org website: organism, gene, description
etc.
"""
seqids = get_all_seqids(protxml_groups)
if is_url_connected('http://uniprot.org'):
uniprot_dict = uniprot.get_metadata_with_some_seqid_conversions(seqids, cache_file)
else:
print "Can't connect to www.uniprot.org, won't use uniprot metatdata"
uniprot_dict = {}
for group_id, protxml_group in protxml_groups.items():
for protein in protxml_group['proteins']:
protein['id'] = ''
protein['acc'] = protein['protein_name']
names = [protein['protein_name']] + protein['other_seqids']
new_seqid = uniprot.sort_seqids_by_uniprot(names, uniprot_dict)[0]
if new_seqid != protein['protein_name']:
print "Protein group %s%s is better represented with %s than %s" % \
(group_id,
protein['group_sibling_id'],
uniprot.get_naked_seqid(new_seqid),
uniprot.get_naked_seqid(protein['protein_name']))
protein['protein_name'] = new_seqid
protein['other_seqids'] = names[1:]
protein['acc'] = new_seqid
protein['other_seqids'] = ';'.join(protein['other_seqids'])
if new_seqid not in uniprot_dict:
print "No uniprot metadata for protein group %s%s seqid %s" % \
(group_id,
protein['group_sibling_id'],
uniprot.get_naked_seqid(new_seqid))
continue
protein['link'] = ''
uniprot_entry = uniprot_dict[new_seqid]
protein['id'] = uniprot_entry['id']
protein['acc'] = uniprot_entry['accs'][0]
protein['link'] = \
'=HYPERLINK("http://uniprot.org/uniprot/%s")' % \
uniprot_dict[new_seqid]['id']
if 'gene' in uniprot_entry:
protein['gene'] = uniprot_entry['gene']
if 'organism' in uniprot_entry:
protein['organism'] = uniprot_entry['organism']
protein['description'] = '; '.join(uniprot_entry['descriptions'])
if 'length' in uniprot_entry:
protein['length'] = uniprot_entry['length']
pairs = uniprot.batch_uniprot_id_mapping_pairs('P_REFSEQ_AC', 'ACC', seqids)
pprint.pprint(pairs, indent=2)
# Example 2 - get UniProt metadata
uniprot_seqids = [j for i, j in pairs]
uniprot_data = uniprot.batch_uniprot_metadata(uniprot_seqids, 'cache')
pprint.pprint(uniprot_data, indent=2)
# Example 3 - parse for isoforms in metadata
text = open('cache/metadata.0.txt').read()
uniprot_data = uniprot.parse_isoforms(text)
pprint.pprint(uniprot_data)
# Example 4 - chaining commands to map seqids
seqids = "EFG_MYCA1 YP_885981.1 ENSG00000196176 Q91ZU6-8".split()
uniprot_data = uniprot.get_metadata_with_some_seqid_conversions(
seqids, 'cache2')
pprint.pprint(uniprot_data, indent=2)
# Example 4 - chaining commands to map seqids
seqids = "EFG_MYCA1 YP_885981.1 ENSG00000196176 Q91ZU6-8".split()
uniprot_data = uniprot.get_metadata_with_some_seqid_conversions(
seqids, 'cache2')
fasta_db = "test-isoform/Q91ZU6-8.fasta"
read_seqids, fastas = uniprot.read_fasta(fasta_db)
test_sequence = list(fastas.values())[0]['sequence']
print(test_sequence == uniprot_data['Q91ZU6-8']['sequence'])
# Example 5 - check isoforms
seqids = ["Q91ZU6-{}".format(i) for i in [1, 2, 3, 4, 5, 6, 8]]
txt = open('test-isoform/Q91ZU6.txt').read()
results = uniprot.parse_uniprot_metadata_with_seqids(seqids, txt)
pprint.pprint(pairs, indent=2)
# Example 2 - get UniProt metadata
uniprot_seqids = [j for i,j in pairs]
uniprot_data = uniprot.batch_uniprot_metadata(
uniprot_seqids, 'cache')
pprint.pprint(uniprot_data, indent=2)
# Example 3 - parse for isoforms in metadata
text = open('cache/metadata.0.txt').read()
uniprot_data = uniprot.parse_isoforms(text)
pprint.pprint(uniprot_data)
# Example 4 - chaining commands to map seqids
seqids = "EFG_MYCA1 YP_885981.1 ENSG00000196176 Q91ZU6-8".split()
uniprot_data = uniprot.get_metadata_with_some_seqid_conversions(
seqids, 'cache2')
pprint.pprint(uniprot_data, indent=2)
# Example 4 - chaining commands to map seqids
seqids = "EFG_MYCA1 YP_885981.1 ENSG00000196176 Q91ZU6-8".split()
uniprot_data = uniprot.get_metadata_with_some_seqid_conversions(
seqids, 'cache2')
fasta_db = "test-isoform/Q91ZU6-8.fasta"
read_seqids, fastas = uniprot.read_fasta(fasta_db)
test_sequence = list(fastas.values())[0]['sequence']
print(test_sequence == uniprot_data['Q91ZU6-8']['sequence'])
# Example 5 - check isoforms
seqids = ["Q91ZU6-{}".format(i) for i in [1, 2, 3, 4, 5, 6, 8]]
txt = open('test-isoform/Q91ZU6.txt').read()
results = uniprot.parse_uniprot_metadata_with_seqids(seqids, txt)
