def test_ncbiquery(self):
ncbi = NCBITaxa(dbfile=DATABASE_PATH)
id2name = ncbi.get_taxid_translator(['9606', '7507'])
self.assertEqual(id2name[7507], 'Mantis religiosa')
self.assertEqual(id2name[9606], 'H**o sapiens')
name2id = ncbi.get_name_translator(['Mantis religiosa', 'h**o sapiens'])
self.assertEqual(name2id['Mantis religiosa'], [7507])
self.assertEqual(name2id['h**o sapiens'], [9606])
name2id = ncbi.get_name_translator(['Bacteria'])
self.assertEqual(set(name2id['Bacteria']), set([2, 629395]))
out = ncbi.get_descendant_taxa("9605", intermediate_nodes=True)
#Out[9]: [1425170, 741158, 63221, 9606]
self.assertEqual(set(out), set([1425170, 741158, 63221, 9606]))
out = ncbi.get_descendant_taxa("9605", intermediate_nodes=False)
#Out[10]: [1425170, 741158, 63221]
self.assertEqual(set(out), set([1425170, 741158, 63221]))
out = ncbi.get_descendant_taxa("9605", intermediate_nodes=False, rank_limit="species")
#Out[11]: [9606, 1425170]
self.assertEqual(set(out), set([9606, 1425170]))
def run(args):
# add lineage profiles/stats
import re
from ete2 import PhyloTree, NCBITaxa
# dump tree by default
if not args.tree and not args.info and not args.descendants:
args.tree = True
ncbi = NCBITaxa()
all_taxids = {}
all_names = set()
queries = []
if not args.search:
log.error('Search terms should be provided (i.e. --search) ')
sys.exit(-1)
for n in args.search:
queries.append(n)
try:
all_taxids[int(n)] = None
except ValueError:
all_names.add(n.strip())
# translate names
name2tax = ncbi.get_name_translator(all_names)
all_taxids.update([(v, None) for v in name2tax.values()])
not_found_names = all_names - set(name2tax.keys())
if args.fuzzy and not_found_names:
log.warn("%s unknown names", len(not_found_names))
for name in not_found_names:
# enable extension loading
tax, realname, sim = ncbi.get_fuzzy_name_translation(name, args.fuzzy)
if tax:
all_taxids[tax] = None
name2tax[name] = tax
name2realname[name] = realname
name2score[name] = "Fuzzy:%0.2f" %sim
if not_found_names:
log.warn("[%s] could not be translated into taxids!" %','.join(not_found_names))
if args.tree:
if len(all_taxids) == 1:
target_taxid = all_taxids.keys()[0]
log.info("Dumping NCBI descendants tree for %s" %(target_taxid))
t = ncbi.get_descendant_taxa(target_taxid, collapse_subspecies=args.collapse_subspecies, rank_limit=args.rank_limit, return_tree=True)
else:
log.info("Dumping NCBI taxonomy of %d taxa..." %(len(all_taxids)))
t = ncbi.get_topology(all_taxids.keys(),
intermediate_nodes=args.full_lineage,
rank_limit=args.rank_limit,
collapse_subspecies=args.collapse_subspecies)
id2name = ncbi.get_taxid_translator([n.name for n in t.traverse()])
for n in t.traverse():
n.add_features(taxid=n.name)
n.add_features(sci_name=str(id2name.get(int(n.name), "?")))
n.name = "%s - %s" %(id2name.get(int(n.name), n.name), n.name)
lineage = ncbi.get_lineage(n.taxid)
n.add_features(named_lineage = '|'.join(ncbi.translate_to_names(lineage)))
dump(t, features=["taxid", "name", "rank", "bgcolor", "sci_name",
"collapse_subspecies", "named_lineage"])
elif args.descendants:
log.info("Dumping NCBI taxonomy of %d taxa..." %(len(all_taxids)))
print '# ' + '\t'.join(["Taxid", "Sci.Name", "Rank", "descendant_taxids", "descendant_names"])
translator = ncbi.get_taxid_translator(all_taxids)
ranks = ncbi.get_rank(all_taxids)
for taxid in all_taxids:
descendants = ncbi.get_descendant_taxa(taxid, collapse_subspecies=args.collapse_subspecies, rank_limit=args.rank_limit)
print '\t'.join([str(taxid), translator.get(taxid, taxid), ranks.get(taxid, ''),
'|'.join(map(str, descendants)),
'|'.join(map(str, ncbi.translate_to_names(descendants)))])
elif args.info:
print '# ' + '\t'.join(["Taxid", "Sci.Name", "Rank", "Named Lineage", "Taxid Lineage"])
translator = ncbi.get_taxid_translator(all_taxids)
ranks = ncbi.get_rank(all_taxids)
for taxid, name in translator.iteritems():
lineage = ncbi.get_lineage(taxid)
named_lineage = ','.join(ncbi.translate_to_names(lineage))
lineage_string = ','.join(map(str, lineage))
print '\t'.join([str(taxid), name, ranks.get(taxid, ''), named_lineage, lineage_string])
fasta_dict=pickle.load( open( "int_data/fasta_dict.p", "rb" )) #Sequences
#2. Filter dataframe by histone variant
#################
f_hist_df=hist_df[(hist_df['hist_var']=='canonical_H2B')]
#3. Select one variant per taxid
#################
f_hist_df=f_hist_df.drop_duplicates(['taxid','hist_var'])
#4. Filter by list of taxonomy clades
################
parent_nodes=[9443] #131567 - cellular organisms
taxids=list()
for i in parent_nodes:
taxids.extend(ncbi.get_descendant_taxa(i))
f_hist_df=f_hist_df[f_hist_df['taxid'].isin(taxids)]
#5. Take one representative per species or specific rank.
################
#Common ranks: superorder-order-suborder-infraorder-parvorder-superfamily-family-subfamily-genus-species-subspecies
seqtaxids=list(f_hist_df['taxid']) #old list
new_seqtaxids=subsample_taxids(seqtaxids,rank='family') #new subsampled list
f_hist_df=f_hist_df[f_hist_df['taxid'].isin(new_seqtaxids)] #remake the dataframe
#---------------
#Output tree before subsampline
tree = ncbi.get_topology(seqtaxids,intermediate_nodes=False)
print tree.get_ascii(attributes=["sci_name", "rank","taxid"])
#Output after subsampling
def run(args):
# add lineage profiles/stats
import re
from ete2 import PhyloTree, NCBITaxa
# dump tree by default
if not args.tree and not args.info and not args.descendants:
args.tree = True
ncbi = NCBITaxa()
all_taxids = {}
all_names = set()
queries = []
if not args.search:
log.error('Search terms should be provided (i.e. --search) ')
sys.exit(-1)
for n in args.search:
queries.append(n)
try:
all_taxids[int(n)] = None
except ValueError:
all_names.add(n.strip())
# translate names
name2tax = ncbi.get_name_translator(all_names)
for tids in name2tax.values():
for tid in tids:
all_taxids[tid] = None
not_found_names = all_names - set(name2tax.keys())
if args.fuzzy and not_found_names:
log.warn("%s unknown names", len(not_found_names))
for name in not_found_names:
# enable extension loading
tax, realname, sim = ncbi.get_fuzzy_name_translation(name, args.fuzzy)
if tax:
all_taxids[tax] = None
name2tax[name] = [tax]
name2realname[name] = realname
name2score[name] = "Fuzzy:%0.2f" %sim
if not_found_names:
log.warn("[%s] could not be translated into taxids!" %','.join(not_found_names))
if args.tree:
if len(all_taxids) == 1:
target_taxid = all_taxids.keys()[0]
log.info("Dumping NCBI descendants tree for %s" %(target_taxid))
t = ncbi.get_descendant_taxa(target_taxid, collapse_subspecies=args.collapse_subspecies, rank_limit=args.rank_limit, return_tree=True)
else:
log.info("Dumping NCBI taxonomy of %d taxa..." %(len(all_taxids)))
t = ncbi.get_topology(all_taxids.keys(),
intermediate_nodes=args.full_lineage,
rank_limit=args.rank_limit,
collapse_subspecies=args.collapse_subspecies)
id2name = ncbi.get_taxid_translator([n.name for n in t.traverse()])
for n in t.traverse():
n.add_features(taxid=n.name)
n.add_features(sci_name=str(id2name.get(int(n.name), "?")))
n.name = "%s - %s" %(id2name.get(int(n.name), n.name), n.name)
lineage = ncbi.get_lineage(n.taxid)
n.add_features(named_lineage = '|'.join(ncbi.translate_to_names(lineage)))
dump(t, features=["taxid", "name", "rank", "bgcolor", "sci_name",
"collapse_subspecies", "named_lineage"])
elif args.descendants:
log.info("Dumping NCBI taxonomy of %d taxa..." %(len(all_taxids)))
print '# ' + '\t'.join(["Taxid", "Sci.Name", "Rank", "descendant_taxids", "descendant_names"])
translator = ncbi.get_taxid_translator(all_taxids)
ranks = ncbi.get_rank(all_taxids)
for taxid in all_taxids:
descendants = ncbi.get_descendant_taxa(taxid, collapse_subspecies=args.collapse_subspecies, rank_limit=args.rank_limit)
print '\t'.join([str(taxid), translator.get(taxid, taxid), ranks.get(taxid, ''),
'|'.join(map(str, descendants)),
'|'.join(map(str, ncbi.translate_to_names(descendants)))])
elif args.info:
print '# ' + '\t'.join(["Taxid", "Sci.Name", "Rank", "Named Lineage", "Taxid Lineage"])
translator = ncbi.get_taxid_translator(all_taxids)
ranks = ncbi.get_rank(all_taxids)
for taxid, name in translator.iteritems():
lineage = ncbi.get_lineage(taxid)
named_lineage = ','.join(ncbi.translate_to_names(lineage))
lineage_string = ','.join(map(str, lineage))
print '\t'.join([str(taxid), name, ranks.get(taxid, ''), named_lineage, lineage_string])
fasta_dict=pickle.load( open( "int_data/fasta_dict.p", "rb" )) #Sequences
#2. Filter dataframe by histone variant
#################
f_hist_df=hist_df[(hist_df['hist_var']=='canonical_H2B')]
#3. Select one variant per taxid
#################
f_hist_df=f_hist_df.drop_duplicates(['taxid','hist_var'])
#4. Filter by list of taxonomy clades
################
parent_nodes=[9443] #131567 - cellular organisms
taxids=list()
for i in parent_nodes:
taxids.extend(ncbi.get_descendant_taxa(i))
f_hist_df=f_hist_df[f_hist_df['taxid'].isin(taxids)]
#5. Take one representative per species or specific rank.
################
#Common ranks: superorder-order-suborder-infraorder-parvorder-superfamily-family-subfamily-genus-species-subspecies
seqtaxids=list(f_hist_df['taxid']) #old list
new_seqtaxids=subsample_taxids(seqtaxids,rank='family') #new subsampled list
f_hist_df=f_hist_df[f_hist_df['taxid'].isin(new_seqtaxids)] #remake the dataframe
#---------------
#Output tree before subsampline
tree = ncbi.get_topology(seqtaxids,intermediate_nodes=False)
print tree.get_ascii(attributes=["sci_name", "rank","taxid"])
#Output after subsampling
