diff options
Diffstat (limited to 'workflows/pull-data/genbank')
| -rwxr-xr-x | workflows/pull-data/genbank/genbank-fetch-ids.py | 1 | ||||
| -rw-r--r-- | workflows/pull-data/genbank/genbank.py | 3 | ||||
| -rwxr-xr-x | workflows/pull-data/genbank/transform-genbank-xml2yamlfa.py | 78 | ||||
| -rwxr-xr-x | workflows/pull-data/genbank/update-from-genbank.py | 25 | ||||
| -rw-r--r-- | workflows/pull-data/genbank/utils.py | 22 |
5 files changed, 71 insertions, 58 deletions
diff --git a/workflows/pull-data/genbank/genbank-fetch-ids.py b/workflows/pull-data/genbank/genbank-fetch-ids.py index cb48cd8..e9e7315 100755 --- a/workflows/pull-data/genbank/genbank-fetch-ids.py +++ b/workflows/pull-data/genbank/genbank-fetch-ids.py @@ -8,7 +8,6 @@ import argparse import sys -from datetime import date from Bio import Entrez diff --git a/workflows/pull-data/genbank/genbank.py b/workflows/pull-data/genbank/genbank.py index 85d615c..026c03f 100644 --- a/workflows/pull-data/genbank/genbank.py +++ b/workflows/pull-data/genbank/genbank.py @@ -111,7 +111,8 @@ def get_metadata(id, gbseq): # print(n,file=sys.stderr) if n != 'Unpublished': institute,address = n.split(',',1) - submitter.submitter_name = institute.split(') ')[1] + if ")" in institute: + submitter.submitter_name = institute.split(')')[1] submitter.submitter_address = address.strip() except AttributeError: pass diff --git a/workflows/pull-data/genbank/transform-genbank-xml2yamlfa.py b/workflows/pull-data/genbank/transform-genbank-xml2yamlfa.py index 9414864..1a8035d 100755 --- a/workflows/pull-data/genbank/transform-genbank-xml2yamlfa.py +++ b/workflows/pull-data/genbank/transform-genbank-xml2yamlfa.py @@ -33,43 +33,47 @@ states = {} for xmlfn in args.files: print(f"--- Reading {xmlfn}") - with gzip.open(xmlfn, 'r') as f: - xml = f.read().decode() - tree = ET.fromstring(xml) - for gb in tree.findall('./GBSeq'): - valid = None - error = None - meta = {} - id = gb.find("GBSeq_locus").text - basename = dir+"/"+id - print(f" parsing {id}") - try: - valid,meta = genbank.get_metadata(id,gb) - if valid: - # --- write JSON - jsonfn = basename + ".json" - with open(jsonfn, 'w') as outfile: - print(f" writing {jsonfn}") - json.dump(meta, outfile, indent=4) - # --- write FASTA - fa = basename+".fa" - seq = genbank.get_sequence(id,gb) - print(f" writing {fa}") - with open(fa,"w") as f2: - f2.write(f"> {id}\n") - f2.write(seq) - # print(seq) - except genbank.GBError as e: - error = f"{e} for {id}" - print(error,file=sys.stderr) - valid = False - state = {} - state['valid'] = valid - if error: - state['error'] = error - if meta['warnings']: - state['warnings'] = meta['warnings'] - states[id] = state + try: + with gzip.open(xmlfn, 'r') as f: + xml = f.read().decode() + except Exception: + with open(xmlfn, 'r') as f: + xml = f.read() + tree = ET.fromstring(xml) + for gb in tree.findall('./GBSeq'): + valid = None + error = None + meta = {} + id = gb.find("GBSeq_locus").text + basename = dir+"/"+id + print(f" parsing {id}") + try: + valid,meta = genbank.get_metadata(id,gb) + if valid: + # --- write JSON + jsonfn = basename + ".json" + with open(jsonfn, 'w') as outfile: + print(f" writing {jsonfn}") + json.dump(meta, outfile, indent=4) + # --- write FASTA + fa = basename+".fa" + seq = genbank.get_sequence(id,gb) + print(f" writing {fa}") + with open(fa,"w") as f2: + f2.write(f"> {id}\n") + f2.write(seq) + # print(seq) + except genbank.GBError as e: + error = f"{e} for {id}" + print(error,file=sys.stderr) + valid = False + state = {} + state['valid'] = valid + if error: + state['error'] = error + if meta['warnings']: + state['warnings'] = meta['warnings'] + states[id] = state statefn = dir + '/state.json' with open(statefn, 'w') as outfile: diff --git a/workflows/pull-data/genbank/update-from-genbank.py b/workflows/pull-data/genbank/update-from-genbank.py index dca5563..95f5a93 100755 --- a/workflows/pull-data/genbank/update-from-genbank.py +++ b/workflows/pull-data/genbank/update-from-genbank.py @@ -14,22 +14,21 @@ import sys from utils import chunks from Bio import Entrez -Entrez.email = 'another_email@gmail.com' # FIXME -BATCH=100 +Entrez.email = 'another_email@gmail.com' # FIXME + +BATCH = 100 parser = argparse.ArgumentParser() -parser.add_argument('--max', type=int, help='Max queries', required=False) parser.add_argument('--ids', type=str, help='File with ids to fetch, 1 id per line', required=True) parser.add_argument('--out', type=str, help='Directory to write to', required=True) +parser.add_argument('--max', type=int, help='Max queries', required=False) args = parser.parse_args() ids = set() with open(args.ids) as f: - content = f.readlines() - for line in content: - ids.add(line.strip()) + ids.update([line.strip() for line in f]) dir = args.out if not os.path.exists(dir): @@ -37,12 +36,14 @@ if not os.path.exists(dir): request_num = BATCH if args.max: - request_num = min(BATCH,args.max) + request_num = min(BATCH, args.max) + +for num_chunk, ids_chunk in enumerate(chunks(list(ids), request_num)): + xmlfn = os.path.join(dir, f"metadata_{num_chunk}.xml.gz") + print(f"Fetching {xmlfn} ({num_chunk * request_num})", file=sys.stderr) -for i, idsx in enumerate(chunks(list(ids), request_num)): - xmlfn = os.path.join(dir, f"metadata_{i}.xml.gz") - print(f"Fetching {xmlfn} ({i*request_num})",file=sys.stderr) with gzip.open(xmlfn, 'w') as f: - f.write((Entrez.efetch(db='nuccore', id=idsx, retmode='xml').read()).encode()) - if args.max and i*request_num >= args.max: + f.write(Entrez.efetch(db='nuccore', id=ids_chunk, retmode='xml').read().encode()) + + if args.max and num_chunk * request_num >= args.max: break diff --git a/workflows/pull-data/genbank/utils.py b/workflows/pull-data/genbank/utils.py index 3efc67a..96920a5 100644 --- a/workflows/pull-data/genbank/utils.py +++ b/workflows/pull-data/genbank/utils.py @@ -1,5 +1,6 @@ import os + def is_integer(string_to_check): try: int(string_to_check) @@ -7,19 +8,26 @@ def is_integer(string_to_check): except ValueError: return False + def chunks(lst, n): for i in range(0, len(lst), n): yield lst[i:i + n] + def check_and_get_ontology_dictionaries(dir_ontology_dictionaries): - # Check duplicated entry looking at all dictionaries + """ + Check duplicated entry by looking in all dictionaries + """ + field_to_term_to_uri_dict = {} - path_dict_xxx_csv_list = [os.path.join(dir_ontology_dictionaries, name_xxx_csv) for name_xxx_csv in - os.listdir(dir_ontology_dictionaries) if name_xxx_csv.endswith('.csv')] + path_dict_xxx_csv_list = [ + os.path.join(dir_ontology_dictionaries, name_xxx_csv) for name_xxx_csv in + os.listdir(dir_ontology_dictionaries) if name_xxx_csv.endswith('.csv') + ] for path_dict_xxx_csv in path_dict_xxx_csv_list: - print('Read {}'.format(path_dict_xxx_csv)) + print(f'Read {path_dict_xxx_csv}') with open(path_dict_xxx_csv) as f: for line in f: @@ -31,7 +39,7 @@ def check_and_get_ontology_dictionaries(dir_ontology_dictionaries): term = term.strip('"') if term in field_to_term_to_uri_dict: - print('Warning: in the dictionaries there are more entries for the same term ({}).'.format(term)) + print(f'Warning: in the dictionaries there are more entries for the same term ({term}).') continue field_to_term_to_uri_dict[term] = uri @@ -54,9 +62,9 @@ def check_and_get_ontology_dictionaries(dir_ontology_dictionaries): term = term.strip('"') if term in field_to_term_to_uri_dict[field]: - print('Warning: in the {} dictionary there are more entries for the same term ({}).'.format(field, term)) + print(f'Warning: in the {field} dictionary there are more entries for the same term ({term}).') continue field_to_term_to_uri_dict[field][term] = uri - return field_to_term_to_uri_dict
\ No newline at end of file + return field_to_term_to_uri_dict |