def test_getRankedDescendants(self):
"""NcbiTaxonNode getRankedDescendants should return correct list"""
nested_species = '''3\t|\t3\t|\tsuperkingdom\t|\t\t|\t8\t|\t0\t|\t1\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
11\t|\t3\t|\tkingdom\t|\t\t|\t8\t|\t0\t|\t1\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
22\t|\t11\t|\tclass\t|\t\t|\t8\t|\t0\t|\t1\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
44\t|\t22\t|\torder\t|\t\t|\t0\t|\t0\t|\t11\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
66\t|\t22\t|\torder\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t|\t
77\t|\t66\t|\tfamily\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
99\t|\t66\t|\tfamily\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
88\t|\t44\t|\tfamily\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
101\t|\t77\t|\tgenus\t|\t\t|\t8\t|\t0\t|\t1\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
202\t|\t77\t|\tgenus\t|\t\t|\t0\t|\t0\t|\t11\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
606\t|\t99\t|\tgenus\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t|\t
707\t|\t88\t|\tgenus\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
909\t|\t88\t|\tgenus\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
123\t|\t909\t|\tgroup\t|\t\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
1111\t|\t123\t|\tspecies\t|\tAT\t|\t8\t|\t0\t|\t1\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
2222\t|\t707\t|\tspecies\t|\tTT\t|\t0\t|\t0\t|\t11\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|
6666\t|\t606\t|\tspecies\t|\tGG\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t|\t
7777\t|\t606\t|\tspecies\t|\tAC\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
9999\t|\t202\t|\tspecies\t|\tBA\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
1010\t|\t101\t|\tspecies\t|\tAC\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
5555\t|\t555\t|\tspecies\t|\tAC\t|\t0\t|\t1\t|\t11\t|\t1\t|\t0\t|\t1\t|\t0\t|\t0\t|\t\t|
555\t|\t3\t|\tsuperclass\t|\t\t|\t8\t|\t0\t|\t1\t|\t0\t|\t0\t|\t0\t|\t0\t|\t0\t|\t\t|'''.split(
'\n')
nested_names = [
'3|a||scientific name|',
'11|b||scientific name|',
'555|c||scientific name|',
'22|d||scientific name|',
'44|e||scientific name|',
'66|f||scientific name|',
'88|g||scientific name|',
'77|h||scientific name|',
'99|i||scientific name|',
'707|j||scientific name|',
'909|k||scientific name|',
'101|l||scientific name|',
'202|m||scientific name|',
'606|n||scientific name|',
'2222|o||scientific name|',
'123|p||scientific name|',
'1111|q||scientific name|',
'1010|r||scientific name|',
'9999|s||scientific name|',
'7777|t||scientific name|',
'6666|u||scientific name|',
'5555|z||scientific name|',
]
tx = NcbiTaxonomyFromFiles(nested_species, nested_names)
dec = tx[3].getRankedDescendants('superclass')
self.assertEqual(len(dec), 1)
assert dec[0] is tx[555]
sp = tx['f'].getRankedDescendants('species')
self.assertSameItems(sp, [tx[1010], tx[9999], tx[7777], tx[6666]])
empty = tx[11].getRankedDescendants('superclass')
self.assertEqual(empty, [])
gr = tx[3].getRankedDescendants('group')
self.assertEqual(gr, [tx[123]])
assert tx[3] is tx['a']
def test_init_bad(self):
"""NcbiTaxonomyFromFiles should produce deadbeats by default"""
bad_tx = NcbiTaxonomyFromFiles(bad_nodes, good_names)
self.assertEqual(len(bad_tx.Deadbeats), 2)
assert 777 in bad_tx.Deadbeats
assert 666 in bad_tx.Deadbeats
assert bad_tx.Deadbeats[777] == bad_tx[9]
def main():
args.infile_nodesdmp_path, args.infile_namesdmp_path, args.infile_mergeddmp_path, \
args.infile_delnodesdmp_path = input_files()
args.outfile_path, args.logfile_path = output_files()
args.output_ranks = args.output_ranks.replace('_', ',').split(',')
accessions_in_input_file = get_accessions_from_input_file()
ncbi_full_taxonomy = NcbiTaxonomyFromFiles(open(args.infile_nodesdmp_path),
open(args.infile_namesdmp_path))
merged_taxids = get_merged_nodes()
deleted_taxids = get_deleted_nodes()
included_nodes, taxid, missing_accessions = obtain_nodes_for_each_accession(accessions_in_input_file, \
args.infile_acc2taxid_path, ncbi_full_taxonomy, merged_taxids, deleted_taxids)
taxid_taxonomy, missing_taxonomy = generate_taxonid_taxonomy(
included_nodes, ncbi_full_taxonomy, args.output_ranks)
generate_output_files(args.outfile_path, taxid_taxonomy, taxid,
missing_accessions, missing_taxonomy)
log_file = open(args.logfile_path, 'a')
log_file.write('Finished running entrez_qiime.py at ' +
strftime("%H:%M:%S on %d-%m-%Y", localtime()) + '\n')
log_file.close()
def get_tax_local(taxid_dict, ranks):
"""
Fetch complete lineage locally (BETTER for huge queries)
Return a dict with:
k: GI
v: complete lineage
"""
from cogent.parse.ncbi_taxonomy import NcbiTaxonomyFromFiles
tree = NcbiTaxonomyFromFiles(open( DB_PATH + 'nodes.dmp'), open( DB_PATH + 'names.dmp'))
root = tree.Root
def get_lineage(node, my_ranks):
ranks_lookup = dict([(r,idx) for idx,r in enumerate(my_ranks)])
lineage = [None] * len(my_ranks)
curr = node
while curr.Parent is not None:
if curr.Rank in ranks_lookup:
lineage[ranks_lookup[curr.Rank]] = curr.Name
curr = curr.Parent
return lineage
tax_dict = {}
for gi in taxid_dict:
# Get lineage for each (gi i.e taxid)
try:
node = tree.ById[taxid_dict[gi]]
tax_dict[gi] = get_lineage(node, ranks)
except KeyError:
print "Cannot Fetch taxonomy for GI: " + gi
print "DONE: Complete taxonomy retrived for each GI"
return tax_dict
def load_ncbi_tree(ncbi_nodes_file, ncbi_names_file):
"""Loading NCBI taxnomy tree
"""
try:
ncbi_tree = NcbiTaxonomyFromFiles(open(ncbi_nodes_file, "rt"),
open(ncbi_names_file, "rt"))
except IOError:
sys.exit("Error cannot open {0} or {1}".format(ncbi_nodes_file,
ncbi_names_file))
return ncbi_tree
def test_init_strict(self):
"""NcbiTaxonomyFromFiles should fail if strict and deadbeats exist"""
tx = NcbiTaxonomyFromFiles(good_nodes, good_names, strict=True)
self.assertRaises(MissingParentError, NcbiTaxonomyFromFiles, \
bad_nodes, good_names, strict=True)
def setUp(self):
self.tx = NcbiTaxonomyFromFiles(good_nodes, good_names)
#!/usr/bin/python2
import sys
import re
## RUN FROM INFO115
from cogent.parse.ncbi_taxonomy import NcbiTaxonomyFromFiles
tree = NcbiTaxonomyFromFiles(open('/2/scratch/brianne/taxonomy/nodes.dmp'),
open('/2/scratch/brianne/taxonomy/names.dmp'))
root = tree.Root
# Here is an example of how to get the NCBI taxon IDs for all Eukaryota phylums
# euks = root.getNodeMatchingName('Eukaryota')
# for node in euks.getRankedDescendants('phylum'):
# print node.Name, node.TaxonId
#
# Chlorophyta 3041
# Streptophyta 35493
# Chytridiomycota 4761
# Microsporidia 6029
# Glomeromycota 214504
# Neocallimastigomycota 451455
# Blastocladiomycota 451459
# Ascomycota 4890
#
# etc
# to get the full lineage for defined ranks based on an input Taxon id
ranks = [
'superkingdom', 'phylum', 'class', 'order', 'family', 'genus', 'species'
]
NCBI_DB + '\" | tee -a blast_taxid.txt',
shell=True)
# New code ends
map_dict = {}
with open('blast_taxid.txt') as gi_taxid:
for each in gi_taxid:
each = each.split()
gi = each[0]
taxid = each[1]
map_dict[gi] = taxid
print 'reading in files for ' + SAMPLE_ID
# tree = NcbiTaxonomyFromFiles(open('/home/roli/scripts/Python/GI_annotations/nodes.dmp'), open('/home/roli/scripts/Python/GI_annotations/names.dmp'))
tree = NcbiTaxonomyFromFiles(open('nodes.dmp'), open('names.dmp'))
root = tree.Root
print 'generating lineages for ' + SAMPLE_ID
error = open(SAMPLE_ID + '.no.lineage.found.out', 'w')
error.write(header2 + "\n")
if re.search("1|superkingdom", TAX_LEVEL):
output_kingdom = open(SAMPLE_ID + '.superkingdom.blast_lineage.out',
'w')
output_kingdom.write(header + "\n")
if re.search("2|phylum", TAX_LEVEL):
output_phylum = open(SAMPLE_ID + '.phylum.blast_lineage.out', 'w')
output_phylum.write(header + "\n")
ranks= [a]
## Define function for pulling lineage info from NCBI nodes and names files
def get_lineage(node, my_ranks):
ranks_lookup = dict([(r,idx) for idx,r in enumerate(my_ranks)])
lineage = [None] * len(my_ranks)
curr = node
while curr.Parent is not None:
if curr.Rank in ranks_lookup:
lineage[ranks_lookup[curr.Rank]] = curr.Name
curr = curr.Parent
return lineage
## IMPORT DIR NAMES into DICTIONARY
tree = NcbiTaxonomyFromFiles(open('/home/roli/db/nodes.dmp'), open('/home/roli/db/names.dmp'))
root = tree.Root
output = open("LINEAGES.tsv", "w")
with open(FILE_LIST) as f:
for line in f:
line = line.strip("\r\n")
line = line.split("\t")
NAME = line[0]
TaxID = int(line[1])
node = tree.ById[TaxID]
tax = get_lineage(node, ranks)
tax = str(tax[0]).lower()