def search_kegg(accessions):
start_time = datetime.datetime.now()
with yaspin(text="Retrieving KEGG annotations...", color="cyan") as sp:
raw_data = ""
for accession in accessions.dropna():
path = KEGG()
res = accession.split(":")
try:
for k, val in path.get_pathway_by_gene(res[1], res[0]).items():
_id = re.search("\d+", k).group(0)
raw_data = f"{raw_data}map{_id}\t\"{val}\"\n"
except AttributeError:
pass
try:
kegg = pandas.read_csv(pandas.compat.StringIO(raw_data),
sep="\t",
header=None)
kegg.columns = ["accession", "description"]
# Add column of counts.
kegg["count"] = kegg.groupby("accession")["accession"].transform(
"count")
kegg = (kegg.drop_duplicates(subset="accession").sort_values(
by="count", ascending=False).reset_index(drop=True))
mssg = f"* Found {sum(kegg['count'])} KEGG pathways from which {len(kegg)} were unique."
except pandas.errors.EmptyDataError:
kegg = pandas.DataFrame()
mssg = f"* Found 0 KEGG Pathways."
time_diff = (datetime.datetime.now() - start_time).total_seconds()
sp.text = f"Retrieving KEGG annotations => Task done in {time_diff} seconds."
sp.ok("✔")
print(mssg)
return kegg
def search_pathway(gene, organism):
k = KEGG()
return k.get_pathway_by_gene(gene, organism)
with open(sys.argv[1], 'r') as fh:
data = fh.read()
data = data.replace('\t', '\n')
inputGenes = data.split('\n')
if '' in inputGenes:
inputGenes.remove('')
#############################
############## Main part
result = {}
for inputGene in inputGenes:
print('Processing gene ' + inputGene + ':')
try:
pathways = k.get_pathway_by_gene(inputGene, 'hsa')
except (AttributeError):
print('Invalid gene identifier')
continue
else:
if pathways:
for pathway in pathways:
if pathway != 'hsa01100':
print('\tProcessing pathway ' + pathway)
rel = k.parse_kgml_pathway(pathway)
genes_result = []
####### Part where the functions are called
for entry in rel['entries']:
if entry['type'] == 'gene':
