def main(infile, ofile, force=False):
ofile = process_path(ofile)
order_id_list, id2annotate = parse_id(infile)
convertor = NCBI_convertor(order_id_list, db='protein')
pid2assembly_dict = convertor.pid2assembly()
if not exists(dirname(ofile)):
os.makedirs(dirname(ofile))
if exists(ofile) and not force:
tqdm.write("detect existing " + ofile +
' no force param input, so it quit instead of writing.')
return
with open(ofile, 'w') as f1:
print('#accession ID\tGI\tassembly_ID\tnuccore ID\tstart\tend\tstrand',
file=f1)
for pid in convertor.origin_ids:
GI = convertor.GI[pid]
_dict = pid2assembly_dict[pid]
assembly_id = _dict['assembly']
info = _dict['nuc_info']
print(f'{pid}\t{GI}\t{assembly_id}\t' + '\t'.join(map(str, info)),
file=f1)
tqdm.write('finish writing into ' + ofile)
def main(infile, ofile, db='protein', force=False, redo=False):
if not exists(dirname(ofile)):
os.makedirs(dirname(ofile))
order_id_list, id2annotate = parse_id(infile)
if db == 'protein':
convertor = NCBI_convertor(order_id_list, db='protein')
pid2assembly_dict = convertor.pid2assembly()
genome_IDs = list(
[_dict['assembly'] for pid, _dict in pid2assembly_dict.items()])
genome_IDs = [_ for _ in genome_IDs if _]
elif db == 'genome':
genome_IDs = order_id_list[::]
else:
raise SyntaxError('wrong input of start_at')
gid2assembly_info, bp2info, bs2info = genomeID2Bio(genome_IDs)
# if not too big, use panda to concat them
# else...... too complicated...pass it
if len(gid2assembly_info) <= 15000:
ginfo_df = pd.DataFrame.from_dict(gid2assembly_info, orient='index')
# ginfo_df.index = ginfo_df.iloc[:,0]
bp_df = pd.DataFrame.from_dict(bp2info, orient='index')
bs_df = pd.DataFrame.from_dict(bs2info, orient='index')
_df1 = bp_df.reindex(ginfo_df.loc[:, 'BioprojectAccn'])
_df1.index = ginfo_df.index
_df2 = bs_df.reindex(ginfo_df.loc[:, 'BioSampleAccn'])
_df2.index = ginfo_df.index
full_df = pd.concat([ginfo_df, _df1, _df2], axis=1)
full_df = full_df.applymap(lambda x: x.replace('\n', ' ')
if isinstance(x, str) else x)
full_df = full_df.drop(['GI', 'relative biosample'], axis=1)
if exists(ofile) and not force:
tqdm.write("detect existing " + ofile +
' no force param input, so it quit instead of writing.')
return
full_df.to_csv(ofile,
sep='\t',
index=1,
index_label='AssemblyAccn raw')
tqdm.write('finish writing into ' + ofile +
' with tab separator format.')
else:
raise Exception('too much genomes to process')
def id2tax(id_list, redo=False, db="protein"):
convertor = NCBI_convertor(id_list, db)
suffix = 'protein2GI'
convertor.check_cache(suffix=suffix, redo=redo)
convertor.get_taxon()
pid2info_dict = defaultdict(dict)
for oid in convertor.origin_ids:
pid2info_dict[oid]['GI'] = convertor.GI[oid]
pid2info_dict[oid]['taxid'] = convertor.tids[oid]
pid2info_dict[oid]['accession'] = oid
access_intermedia(pid2info_dict, suffix=suffix)
return pid2info_dict
def genomeID2Bio(genome_IDs):
"""
bio include BioSample and corresponding BioProject
"""
tqdm.write("with genome ID, start to retrieve genome information")
convertor = NCBI_convertor(genome_IDs, db='assembly')
convertor.get_GI()
convertor.get_db_summary()
aid2info = convertor.dbsummary
bs_list = list(
set([
_.get('BioSampleAccn') for _ in aid2info.values()
if _.get('BioSampleAccn')
]))
bp_list = list(
set([
_.get('BioprojectAccn') for _ in aid2info.values()
if _.get('BioprojectAccn')
]))
tqdm.write("retrieving relative Bioproject and its Biosample info")
bp2info = get_bioproject(bp_list)
bs2info = get_biosample(bs_list)
return aid2info, bp2info, bs2info
def aid2taxon(id_list, redo=False):
convertor = NCBI_convertor(id_list, "assembly")
suffix = 'protein2GI'
convertor.check_cache(suffix=suffix, redo=redo)
id2taxon = convertor.get_taxons_from_tid()
return id2taxon
from bin.ncbi_convertor import NCBI_convertor
if __name__ == "__main__":
# test
pids = open('./protein_ids').read().split('\n')
convertor = NCBI_convertor(pids, db='protein')
# convertor.check_cache(suffix=suffix, redo=redo)
convertor.get_taxons_from_tid()
pid2assembly_dict = convertor.pid2assembly()
aids = open('./assembly_ids').read().split('\n')
convertor = NCBI_convertor(aids, db='assembly')
# convertor.check_cache(suffix=suffix, redo=redo)
convertor.get_taxons_from_tid()
nids = open('./nucleotide_ids').read().split('\n')
convertor = NCBI_convertor(nids, db='nuccore')
convertor.get_GI()
convertor.get_db_summary()
# convertor.check_cache(suffix=suffix, redo=redo)
convertor.get_taxons_from_tid()