def to_html(self, title=None):
from kobas import config
from Cheetah.Template import Template
tmpl = os.path.join(config.getrc()["kobas_home"], "template", "test_html.tmpl")
self.add_link_to_pathway()
t = Template(file=tmpl, searchList=[{"thead": self.title, "result": self.result}])
return str(t)
def __init__(self, dbfile="", *args, **kwargs):
""" init a keggdb object
"""
if dbfile:
self.db = sqlite.connect(dbfile, *args, **kwargs)
else:
self._rc = config.getrc()
self.db = sqlite.connect(self._rc['keggdb'], *args, **kwargs)
self.cursor = self.db.cursor()
def __init__(self, dbfile="", *args, **kwargs):
""" init a keggdb object
"""
if dbfile:
self.db = sqlite.connect(dbfile, *args, **kwargs)
else:
self._rc = config.getrc()
self.db = sqlite.connect(self._rc["keggdb"], *args, **kwargs)
self.cursor = self.db.cursor()
def to_html(self, title=None):
from kobas import config
from Cheetah.Template import Template
tmpl = os.path.join(config.getrc()['kobas_home'], "template",
"test_html.tmpl")
self.add_link_to_pathway()
t = Template(file=tmpl,
searchList=[
{
'thead': self.title,
'result': self.result
},
])
return str(t)
def requires(self):
global ref_fa, genome_dir, ref_gtf, annotation_dir, species_latin, species_ensembl, species_kegg, log_dir, ko_pep_dir, ko_db_dir
ref_fa, ref_gtf, species_latin = self.ref_fa, self.ref_gtf, self.species_latin
species_kegg, species_ensembl = get_kegg_biomart_id(species_latin)
genome_dir = path.dirname(ref_fa)
annotation_dir = path.dirname(ref_gtf)
log_dir = path.join(annotation_dir, 'logs')
kobasrc = config.getrc()
ko_pep_dir = kobasrc['blastdb']
ko_db_dir = kobasrc['kobasdb']
circ_mkdir_unix(log_dir)
print ko_pep_dir
return [fa_index(), star_index(), go_annotation(), ko_annotation()]
def arg2dist(arg):
""" get dist dict in term of arg, abbrev for genome or used defined annot
"""
kobasrc = config.getrc()
if len(arg) == 3:
keggdb = dbutils.keggdb()
distfile = os.path.join(kobasrc["kobas_home"], "gene_dist",
keggdb.get_species_name(arg.lower()))
dist_dict = discover.dist_from_distfile(open(distfile))
else:
if not os.access(arg, os.F_OK):
print "File: %s not exist" % arg
sys.exit(0)
dist_dict = discover.dist_from_annot_file(open(arg, 'r'))
dist_size = dist_dict.size()
return (dist_dict, dist_size)
def distr_from_default(default, db, speciesdb, abbr, small_num, idmapping):
distr = {}
if abbr == 'ko':
koids = speciesdb.kos()
for koid in koids:
terms = speciesdb.pathways_from_koid(koid['koid'])
gids = speciesdb.gids_from_koid(koid['koid'], default)
for gid in gids:
if idmapping:
bspeciesdb = dbutils.KOBASDB(config.getrc()['kobasdb'] + default + '.db')
dblink_ids = [dblink_id[0] for dblink_id in bspeciesdb.dblink_ids_from_gid(gid['gid'], idmapping)]
for term in terms:
if idmapping:
distr[(term['name'], db, term['pid'])] = distr.get((term['name'], db, term['pid']), 0) + len(dblink_ids)
else:
distr[(term['name'], db, term['pid'])] = distr.get((term['name'], db, term['pid']), 0) + 1
else:
gids = speciesdb.genes()
for gid in gids:
if db in dbutils.P.keys():
terms = speciesdb.pathways_from_gid(gid['gid'], db)
elif db in dbutils.D.keys():
terms = speciesdb.diseases_from_gid(gid['gid'], db)
elif db in dbutils.G.keys():
terms = speciesdb.gos_from_gid(gid['gid'])
if idmapping:
dblink_ids = [dblink_id[0] for dblink_id in speciesdb.dblink_ids_from_gid(gid['gid'], idmapping)]
if db in dbutils.P.keys() + dbutils.D.keys():
for term in terms:
if idmapping:
distr[(term['name'], db, term['id'])] = distr.get((term['name'], db, term['id']), 0) + len(dblink_ids)
else:
distr[(term['name'], db, term['id'])] = distr.get((term['name'], db, term['id']), 0) + 1
elif db in dbutils.G.keys():
for term in terms:
if idmapping:
distr[(term['name'], db, term['goid'])] = distr.get((term['name'], db, term['goid']), 0) + len(dblink_ids)
else:
distr[(term['name'], db, term['goid'])] = distr.get((term['name'], db, term['goid']), 0) + 1
remove_small_terms2(distr, small_num)
return distr
def arg2dist(arg):
""" get dist dict in term of arg, abbrev for genome or used defined annot
"""
kobasrc = config.getrc()
if len(arg) == 3:
keggdb = dbutils.keggdb()
distfile = os.path.join(
kobasrc["kobas_home"], "gene_dist",
keggdb.get_species_name(arg.lower()))
dist_dict = discover.dist_from_distfile(open(distfile))
else:
if not os.access(arg, os.F_OK):
print "File: %s not exist" % arg
sys.exit(0)
dist_dict = discover.dist_from_annot_file(open(arg, 'r'))
dist_size = dist_dict.size()
return (dist_dict, dist_size)
import kobas.config as kobas_config
import os
PEP_URL_TEMP = 'ftp://ftp.cbi.pku.edu.cn/pub/KOBAS_3.0_DOWNLOAD/seq_pep/{}.pep.fasta.gz'
DB_URL_TEMP = 'ftp://ftp.cbi.pku.edu.cn/pub/KOBAS_3.0_DOWNLOAD/sqlite3/{}.db.gz'
BLAST_DEFAULT = '-evalue 1e-5 -outfmt 6 -max_target_seqs 1'
KOBASRC = kobas_config.getrc()
BLAST_DIR = KOBASRC['blastout']
PEP_DIR = KOBASRC['blastdb']
DB_DIR = KOBASRC['kobasdb']
KOBAS_RUN = KOBASRC['kobas_run']
KOBAS_PY = KOBASRC['kobas_py']
WHEAT_PEP_DB = os.path.join(BLAST_DIR, 'wheat.pep.fa')
TERM_CAT = os.path.join(BLAST_DIR, 'term.cat.txt')
def read_gmt_db(gene_list, species, dbtype, idtype, min_size, max_size):
gset_name = [] #name of gene sets
gset_des = [] #description of gene sets
hit_genes = [
] #genes in gene sets none of use, only use genes both in list and gene set
D = {
'o': 'OMIM',
'k': 'KEGG DISEASE',
'f': 'FunDO'
} #don't support GAD and N, because lack of did
try:
kobasrc = config.getrc()
speciesdb = dbutils.KOBASDB(kobasrc['kobasdb'] + species + '.db')
dbcatg = dbtype.split(':')[0]
if not dbcatg in ['P', 'D', 'G']:
raise exception.GmtdatabaseErr, "Error Message: This database category you input is not allowed. Please choose 'P','D','G' for database category."
records = []
gset_num = 0
if dbcatg == 'P':
input_dbs = set(dbtype.split(':')[1].split('/'))
avail_dbs = {}
for database in speciesdb.pathwaydbs_from_abbr(species):
avail_dbs[database[0]] = database[1]
dbs = input_dbs.intersection(set(avail_dbs.keys()))
if dbs:
print "Databases: %s" % ', '.join(
[avail_dbs[db] for db in dbs])
else:
raise exception.GmtdatabaseErr, "No supported databases are selected. Supported databases are %s, but your input databases are: %s." % \
('/'.join(avail_dbs.keys()), dbtype)
for db in dbs:
tmp = speciesdb.allpathways(db).fetchall()
num = speciesdb.pathwaynums(db)
records += tmp
gset_num += num
if gset_num == 0:
raise exception.GmtdatabaseErr, "Error Message: \nFail to get information from gene set database."
i = 0
hit_matrix = np.array([[0.0 for s in range(len(gene_list))]
for m in range(gset_num)])
for record in records:
hgene = []
gset_name.append(record[1]) #id
gset_des.append(record[2]) #name
gset_gids = speciesdb.genes_from_pid(
record[0]) #from table GenePathways
for gid in gset_gids:
#print gid
#dblink_id = speciesdb.dblink_ids_from_gid(gid[0], idtype).fetchone()[0]
for each in speciesdb.dblink_ids_from_gid(gid[0], idtype):
dblink_id = each[0]
if dblink_id in gene_list:
hgene.append(dblink_id)
j = gene_list.index(dblink_id)
hit_matrix[i, j] = 1
hit_genes.append(hgene)
i += 1
if not 1 in hit_matrix:
raise exception.GmtdatabaseErr, "Error Message:\nNone of genes in input file(gct file) was matched with genes in gene sets. \nPlease check the gct file(-e), the species(-s), the idtype(-i), the database type(-d)."
elif dbcatg == 'D':
input_dbs = set(dbtype.split(':')[1].split('/'))
avail_dbs = {}
if species != 'hsa':
raise exception.GmtdatabaseErr, "Error Message: Disease is only supported for H**o Sapiens(-s hsa), not supported for this species. Please choose another database category, e.g. 'P','G'"
dbs = input_dbs.intersection(set(D.keys()))
# if dbs:
# print 'Databases: %s' % ', '.join([avail_dbs[db] for db in dbs])
# else:
if not dbs:
raise exception.GmtdatabaseErr, 'No supported databases are selected. Supported databases are %s, but your input databases are: %s.' % \
('/'.join(avail_dbs.keys()), dbtype)
for db in dbs:
tmp = speciesdb.alldiseases(db).fetchall()
num = speciesdb.diseasenums(db)
records += tmp
gset_num += num
if gset_num == 0:
raise exception.GmtdatabaseErr, "Error Message: \nFail to get information from gene set database."
i = 0
hit_matrix = np.array([[0.0 for s in range(len(gene_list))]
for m in range(gset_num)])
for record in records:
hgene = []
#gset_name.append(record[1]) #id
if record[1]:
gset_name.append(record[1])
else:
#gset_name.append(record[2])
continue
gset_des.append(record[2]) #name
gset_gids = speciesdb.genes_from_did(
record[0]) #from table GenePathways
for gid in gset_gids:
#dblink_id = speciesdb.dblink_ids_from_gid(gid[0], idtype).fetchone()[0]
for each in speciesdb.dblink_ids_from_gid(gid[0], idtype):
dblink_id = each[0]
if dblink_id in gene_list:
hgene.append(dblink_id)
j = gene_list.index(dblink_id)
hit_matrix[i, j] = 1
hit_genes.append(hgene)
i += 1
if not 1 in hit_matrix:
raise exception.GmtdatabaseErr, "Error Message:\nNone of genes in input file(gct file) was matched with genes in gene sets. \nPlease check the gct file(-e), the species(-s), the idtype(-i), the database type(-d)."
else:
records = speciesdb.allgoterms().fetchall()
gset_num = speciesdb.gotermnums()
i = 0
hit_matrix = np.array([[0.0 for s in range(len(gene_list))]
for m in range(gset_num)])
for record in records:
hgene = []
gset_name.append(record[0]) #id
gset_des.append(record[1]) #name
gset_gids = speciesdb.genes_from_goid(
record[0]) #from table GenePathways
for gid in gset_gids:
#dblink_id = speciesdb.dblink_ids_from_gid(gid[0], idtype).fetchone()[0]
dblink_id = []
for each in speciesdb.dblink_ids_from_gid(gid[0], idtype):
dblink_id += each[0]
if dblink_id in gene_list:
hgene.append(dblink_id)
j = gene_list.index(dblink_id)
hit_matrix[i, j] = 1
hit_genes.append(hgene)
i += 1
if not 1 in hit_matrix:
raise exception.GmtdatabaseErr, "Error Message:\nNone of genes in input file(gct file) was matched with genes in gene sets. \nPlease check the gct file(-e), the species(-s), the idtype(-i), the database type(-d)."
gset_name = np.array(tuple(gset_name))
gset_des = np.array(gset_des)
hit_genes = np.array(hit_genes)
hitsum = hit_matrix.sum(1)
delindex = np.where((hitsum < min_size) | (hitsum > max_size))[0]
hit_matrix_filtered = np.delete(hit_matrix, delindex, axis=0)
hit_genes_filterd = np.delete(hit_genes, delindex, axis=0)
hit_sum_filtered = np.delete(hitsum, delindex, axis=0)
gset_name_filtered = np.delete(np.array(gset_name), delindex, axis=0)
gset_des_filtered = np.delete(gset_des, delindex, axis=0)
if len(hit_matrix_filtered) == 0 and len(
gset_name_filtered) == 0 and len(
gset_des_filtered) == 0 and len(hit_genes_filterd) == 0:
raise exception.GmtvalueErr, "Error Message:\nAll gene sets" + repr(
len(hit_matrix)
) + "have been filtered. \nPlease check the threshold and ceil of gene set size (values of min_size and max_size). "
except ValueError, e:
sys.exit(e)
for abbr in pathways.keys():
speciesdb = dbutils.KOBASDB(kobasrc['kobasdb'] + abbr + '.db')
count = 0
for pname in pathways[abbr].keys():
count += 1
pathways[abbr][pname][0] = count * 10 + 4
if __name__ == '__main__':
print pathways[abbr][pname][0], 'R', pathways[abbr][pname][
1], pname
else:
speciesdb.con.execute('INSERT INTO Pathways VALUES (?, ?, ?, ?)', \
(pathways[abbr][pname][0], 'R', pathways[abbr][pname][1], pname))
for gene_pathway in gene_pathways[abbr]:
if __name__ == '__main__':
print gene_pathway[0], pathways[abbr][gene_pathway[1]][0]
else:
speciesdb.con.execute('INSERT INTO GenePathways VALUES (?, ?)', \
(gene_pathway[0], pathways[abbr][gene_pathway[1]][0]))
if __name__ == '__main__':
import sys
from kobas import config
parse(open(sys.argv[1]), open(sys.argv[2]), config.getrc())
for pathway in tuple(pathways):
if pathway[0] not in pids:
pathways.remove(pathway)
pathways = tuple(pathways)
gene_pathways = tuple(gene_pathways)
return pathways, gene_pathways
if __name__ == '__main__':
import sys
from pprint import pprint
from kobas import config, dbutils
pathways, gene_pathways = parse(open(sys.argv[1]), \
dbutils.KOBASDB(config.getrc()['kobasdb'] + 'hsa.db'))
print len(pathways), len(gene_pathways)
g, p = set(), set()
for gp in gene_pathways:
g.add(gp[0])
p.add(gp[1])
print len(g), len(p)
pprint(pathways[:5])
pprint(gene_pathways[:5])
help="specify which program to use by blastall, default blastp")
p.add_option(
"-r", "--rank", dest="rank", action="store", type="string",
help="rank cutoff for valid hit from BLAST, default 5")
(opt,args) = p.parse_args()
return (p,opt,args)
if __name__ == "__main__":
opt_parser,opt,args = config_option()
if len(args) != 1:
opt_parser.print_help()
sys.exit(1)
# Blast environment configuration
kobasrc = config.getrc()
# Integrate command line switches into environments
if opt.rank:
kobasrc["rank"] = opt.rank
if opt.evalue:
kobasrc["evalue"] = opt.evalue
keggdb = dbutils.keggdb(kobasrc['kobasdb'])
if opt.intype == "fasta":
# verify fasta file
try:
try:
f = open(args[0])
fasta.verify(f)
#!/usr/bin/env python # -*- coding: ISO-8859-1 -*- # Copyright by Mao Xizeng ([email protected]) # Created: 2005-03-24 13:14:23 # $Id: kpath.py 465 2008-03-04 18:15:07Z lymxz $ __version__ = '$LastChangedRevision: 465 $'.split()[-2] """KEGG pathway mining based on KO """ from glob import glob from os.path import join from kobas.config import getrc from kobas.kgml import kgml from pygraphlib import pygraph, algo, pydot kg = kgml(join(getrc()['dat_dir'], 'data', 'kgml')) def dgraph_from_kgml(): pathways = kg.get_pathways('ot') edges = [] for k, v in pathways.items(): for l in v.get_linked_pathways(): edges.append((k, l.name)) return pygraph.from_list(edges) if __name__ == "__main__": pass
gids = bcids[gene_id]
for gid in gids:
gene_pathways.add((gid, pid))
pathways = list(pathways)
gene_pathways = list(gene_pathways)
return pathways, gene_pathways
if __name__ == '__main__':
import sys
from pprint import pprint
from kobas import config, dbutils
pathways, gene_pathways = parse(open(sys.argv[1]), open(sys.argv[2]), \
dbutils.KOBASDB(config.getrc()['kobasdb'] + sys.argv[3] + '.db'))
print len(pathways), len(gene_pathways)
g, p = set(), set()
for gp in gene_pathways:
g.add(gp[0])
p.add(gp[1])
print len(g), len(p)
pprint(pathways[:5])
pprint(gene_pathways[:5])
def testCheetah(self):
tmpl = os.path.join(config.getrc()['kobas_home'],
"template", "test_html.tmpl")
data = {'thead':range(7), 'result':[range(7) for i in range(10)], 'title':'Test'}
t = Template(file=tmpl, searchList=[data,])
self.failUnless(str(t).find('$') == -1)
gids = [gid[0] for gid in speciesdb.genes(name=False)]
geid_oeids = get_geid_oeids(orthologydir, abbr)
for gid in gids:
geids = speciesdb.ensembl_gene_ids_from_gid(gid)
for geid in geids:
if geid_oeids.has_key(geid[0]):
for oeid, ospecies in geid_oeids[geid[0]]:
ospeciedb = dbutils.KOBASDB(kobasrc['kobasdb'] + ospecies +
'.db')
oids = ospeciedb.gids_from_ensembl_gene_id(oeid)
for oid in oids:
orthologs.add((gid, oid[0]))
return list(orthologs)
if __name__ == '__main__':
import sys
from pprint import pprint
from kobas import config
kobasrc = config.getrc()
orthologydir = kobasrc['kobas_home'] + '/orthology/'
orthologs = parse(kobasrc, orthologydir, 'hsa')
print len(orthologs)
pprint(orthologs[:5])
): ##if 'not' in column 4, discard the annotation
tid = info[12].split('|')[0][
6:] ##if '|' in column 13, only the first organism encodes the gene or gene product
if tid_abbrs.has_key(tid):
abbr = tid_abbrs[tid]
gpid = info[1].split(':')[-1] ##column 2 is DB Object ID
if organism[1] != 'not_need':
uniprotkb_acs = gp2protein.get(gpid, [])
else:
uniprotkb_acs = [gpid]
goid = info[4] ##column 5 is GO ID
for uniprotkb_ac in uniprotkb_acs:
multi_uniprotkb_ac_gos[abbr].add((uniprotkb_ac, goid))
return multi_uniprotkb_ac_gos
if __name__ == '__main__':
import sys
from pprint import pprint
from kobas import config
multi_uniprotkb_ac_gos = parse(
config.getrc()['kobas_home'],
('gene_association.goa_human', 'not_need', '9606', 'hsa'))
print len(multi_uniprotkb_ac_gos)
print len(multi_uniprotkb_ac_gos['hsa'])
pprint(tuple(multi_uniprotkb_ac_gos['hsa'])[:5])
