def createTree(arr):
"""
Create a tree using the provided list of tuples. In the tree, a node may contains multiple
child nodes, but it can only have one parent node.
:param arr: list of tuples (taxid, weight, lineage), where lineage is a list of taxid
:return: Tree instance
"""
taxadb = TaxID()
root = Node(0, 0, _get_name_for_taxid(0, taxadb))
catalog = {}
for bucket in arr:
taxid = int(bucket[0])
name = _get_name_for_taxid(taxid, taxadb)
leaf = Node(taxid, bucket[1], name)
iter = root
for i in range(len(bucket[2])):
lin = bucket[2][i]
if catalog.get(lin) is None:
node = Node(lin, 0, _get_name_for_taxid(lin, taxadb))
iter.addChildNode(node)
catalog[lin] = node
iter = node
else:
# below we check the compatibility of the data
# each node can only have one parent node
eid = catalog[lin].parentNode.taxid
if i == 0:
if eid != root.taxid:
m = "node #{} got two parent nodes #{} and #{}".format(
lin, eid, root.taxid)
raise ValueError(m)
else:
if eid != bucket[2][i - 1]:
m = "node #{} got two parent nodes #{} and #{}".format(
lin, eid, bucket[2][i - 1])
raise ValueError(m)
iter = catalog[lin]
if catalog.get(leaf.taxid) is None:
iter.addChildNode(leaf)
catalog[leaf.taxid] = leaf
else:
eid = catalog[bucket[0]].parentNode.taxid
if eid != bucket[2][-1]:
m = "node #{} got two parent nodes #{} and #{}".format(
bucket[0], eid, bucket[2][-1])
raise ValueError(m)
iter = catalog[bucket[0]]
iter.updateWeight(leaf.weight)
return Tree(root, catalog, root.weight)
def harvest(outputs, taxadb):
queue = []
# process command line args
i = 0
while i < len(outputs):
if outputs[i].endswith("="):
assert i + 1 < len(outputs), f"missing path after {outputs[i]}"
sample = outputs[i][:-1]
rundir = outputs[i + 1]
i += 2
else:
parts = outputs[i].split("=")
assert len(parts) == 2, f"invalid SAMPLE_ID=RUNDIR pair: {outputs[i]}"
sample = parts[0]
rundir = parts[1]
i += 1
assert sample in BENCHMARKS["samples"], f"unknown sample {sample}"
if rundir.startswith("s3:"):
assert os.environ.get(
"AWS_PROFILE", False
), f"set environment AWS_PROFILE to read from {rundir}"
else:
assert (
Path(rundir) / "outputs.json"
).is_file(), f"couldn't find outputs.json in {rundir}"
queue.append((sample, rundir))
if taxadb:
taxadb = TaxID(dbtype="sqlite", dbname=taxadb)
# harvest each supplied sample
rslt = {}
for sample, rundir in queue:
assert sample not in rslt, f"repeated sample {sample}"
outputs_json = read_outputs_json(rundir)
rslt[sample] = harvest_sample(sample, outputs_json, taxadb)
rslt[sample]["outputs"] = outputs_json
print(json.dumps(rslt, indent=2))
def main():
"""
Main program function
"""
# Get arguments
args = get_arguments()
# Step 1
print("STEP 1: Extracting Genbank IDS from BLAST output...")
accession = extract_genbank_id(args.blast_output_file)
chunks = [accession[i:i + 999] for i in range(0, len(accession), 999)]
print("Found {0} ids !".format(len(accession)))
#print(accession)
# Step 3
print("STEP 2: Writing results to file '{0}'...".format(
args.taxonomy_file))
accession_db = AccessionID(dbtype='sqlite', dbname=args.taxadb_file)
tax_db = TaxID(dbtype='sqlite', dbname=args.taxadb_file)
taxids = accession_db.taxid(accession)
#taxids = [accession_db.taxid(acc) for acc in accession]
#print(taxids)
write_results(chunks, accession_db, tax_db, args.taxonomy_file)
print("DONE !")
#!/usr/bin/env python3
import sys
from os.path import expanduser
from taxadb.taxid import TaxID
handler = TaxID(dbtype="sqlite", dbname=expanduser("~") + "/.taxadb/taxadb.sqlite")
for taxid in sys.stdin:
taxid = taxid.strip()
lineage = handler.lineage_name(taxid, reverse=True)
lineage_s = ""
if len(lineage) > 0:
lineage_s = "; ".join(lineage)
print("{}\t{}".format(taxid, lineage_s))
def main():
parser = argparse.ArgumentParser(
description='Filter a Uniref FASTA file by taxonomy')
## output file to be written
parser.add_argument('-i',
'--input_fasta',
type=str,
required=True,
help='Path to an input file to be read')
parser.add_argument('-o',
'--output_fasta',
type=str,
required=True,
help='Path to an output file to be created')
parser.add_argument(
'-c',
'--clades',
type=str,
required=True,
help='Comma-separated string of clades to be included.')
parser.add_argument('-db',
'--taxadb',
type=str,
required=True,
help='Path to the taxadb sqlite3 file')
args = parser.parse_args()
if not os.path.exists(args.taxadb):
raise Exception("ERROR: Couldn't find taxadb specified")
taxid = TaxID(dbtype='sqlite', dbname=args.taxadb)
clades = args.clades.split(',')
record_count = 0
print_every = 1000
clade_counter = dict()
for clade in clades:
clade_counter[clade] = 0
# remembers for each ID if we're keeping it or not
id_cache = dict()
fout = open(args.output_fasta, 'wt')
keep_entry = False
for line in open(args.input_fasta):
if line[0] == '>':
record_count += 1
if record_count % print_every == 0:
print("{0} records processed ...".format(record_count),
flush=True)
m = re.search('TaxID=(\d+)', line)
if m:
tax_id = m.group(1)
if tax_id in id_cache:
if id_cache[tax_id] == True:
keep_entry = True
else:
keep_entry = False
else:
lineage = taxid.lineage_name(tax_id, reverse=True)
clade_found = False
if lineage is None:
keep_entry = False
continue
for clade in clades:
if clade in lineage:
clade_found = True
clade_counter[clade] += 1
break
if clade_found:
keep_entry = True
id_cache[tax_id] = True
else:
keep_entry = False
id_cache[tax_id] = False
else:
keep_entry = False
if keep_entry:
fout.write(line)
fout.close()
print("Number of entries exported by clade:")
for clade in clade_counter:
print("\t{0}: {1}".format(clade, clade_counter[clade]))
#!/usr/bin/env python3
import sys
from taxadb.taxid import TaxID
taxid = TaxID()
for t in sys.argv[1:]:
name = taxid.sci_name(int(t))
print(name)
def main():
parser = argparse.ArgumentParser( description='Filter a Uniref FASTA file by taxonomy')
## output file to be written
parser.add_argument('-i', '--input_fasta', type=str, required=True, help='Path to an input file to be read' )
parser.add_argument('-o', '--output_fasta', type=str, required=True, help='Path to an output file to be created' )
parser.add_argument('-c', '--clades', type=str, required=True, help='Comma-separated string of clades to be included.' )
parser.add_argument('-db', '--taxadb', type=str, required=True, help='Path to the taxadb sqlite3 file' )
args = parser.parse_args()
if not os.path.exists(args.taxadb):
raise Exception("ERROR: Couldn't find taxadb specified")
taxid = TaxID(dbtype='sqlite', dbname=args.taxadb)
clades = args.clades.split(',')
record_count = 0
print_every = 1000
clade_counter = dict()
for clade in clades:
clade_counter[clade] = 0
# remembers for each ID if we're keeping it or not
id_cache = dict()
fout = open(args.output_fasta, 'wt')
keep_entry = False
for line in open(args.input_fasta):
if line[0] == '>':
record_count += 1
if record_count % print_every == 0:
print("{0} records processed ...".format(record_count), flush=True)
m = re.search('TaxID=(\d+)', line)
if m:
tax_id = m.group(1)
if tax_id in id_cache:
if id_cache[tax_id] == True:
keep_entry = True
else:
keep_entry = False
else:
lineage = taxid.lineage_name(tax_id, reverse=True)
clade_found = False
if lineage is None:
keep_entry = False
continue
for clade in clades:
if clade in lineage:
clade_found = True
clade_counter[clade] += 1
break
if clade_found:
keep_entry = True
id_cache[tax_id] = True
else:
keep_entry = False
id_cache[tax_id] = False
else:
keep_entry = False
if keep_entry:
fout.write(line)
fout.close()
print("Number of entries exported by clade:")
for clade in clade_counter:
print("\t{0}: {1}".format(clade, clade_counter[clade]))
# Download TinySeq XML format file from NCBI BioProject 33175:
# https://www.ncbi.nlm.nih.gov/bioproject?db=bioproject&Cmd=DetailsSearch&Term=33175%5Buid%5D
from bs4 import BeautifulSoup as BS
soup = BS(open('./sequence.fasta.xml'), 'lxml')
# You'll need to build the taxa taxadb database
from taxadb.taxid import TaxID
db = TaxID(dbtype='sqlite', dbname='taxadb.sqlite')
import sys
'''
soup.prettify()[:2100] ==
<html>
<body>
<tseqset>
<tseq>
<tseq_seqtype value="nucleotide">
</tseq_seqtype>
<tseq_accver>
NR_170543.1
</tseq_accver>
<tseq_sid>
gnl|REF_SSU16S|KU507537:1-1439
</tseq_sid>
<tseq_taxid>
1849015
</tseq_taxid>
<tseq_orgname>
Pseudoarcobacter acticola
</tseq_orgname>
<tseq_defline>
#!/usr/bin/env python3
from taxadb.taxid import TaxID
import fileinput
taxid = TaxID()
for line in fileinput.input():
print(taxid.sci_name(line.rstrip()))