def organism_version(cls, name):
oname = cls.organism_name_search(name)
#FIXME, dirty hack to ensure file id downloaded
orngServerFiles.localpath_download("NCBI_geneinfo",
"gene_info.%s.db" % oname)
return orngServerFiles.info("NCBI_geneinfo",
"gene_info.%s.db" % oname)["datetime"]
def __init__(self, organism, genematcher=None):
""" An dictionary like object for accessing NCBI gene info
Arguments::
- *organism* Organism id
Example::
>>> info = NCBIGeneInfo("H**o sapiens")
"""
self.taxid = self.organism_name_search(organism)
fname = orngServerFiles.localpath_download("NCBI_geneinfo", "gene_info.%s.db" % self.taxid)
file = open(fname, "rb")
self.update(dict([(line.split("\t", 3)[1], line) for line in file.read().splitlines() if line.strip() and not line.startswith("#")]))
# NOTE orig init time for gene matcher: 2.5s, new 4s: investigate the slowdown
# NOTE matches are not the same because aliases are build a bit
# differently (main name versus old aliases conflict!)
self.matcher = genematcher
if self.matcher == None:
if self.taxid == '352472':
self.matcher = matcher([GMNCBI(self.taxid), GMDicty(), [GMNCBI(self.taxid), GMDicty()]])
else:
self.matcher = matcher([GMNCBI(self.taxid)])
#if this is done with a gene matcher, pool target names
self.matcher.set_targets(self.keys())
def history(self):
if getattr(self, "_history", None) is None:
fname = orngServerFiles.localpath_download("NCBI_geneinfo", "gene_history.%s.db" % self.taxid)
try:
self._history = dict([(line.split("\t")[2], GeneHistory(line)) for line in open(fname, "rb").read().splitlines()])
except Exception, ex:
print >> sys.srderr, "Loading NCBI gene history failed.", ex
self._history = {}
def load(self):
path = orngServerFiles.localpath_download(self.DOMAIN, self.FILENAME)
lines = open(path, "rb").read().splitlines()[:-1]
self._homologs = {}
self._homologs = dict([((h.taxonomy_id, h.gene_symbol), h)
for h in [_homolog(line) for line in lines]])
self._homologs_by_group = reduce(
lambda dict, h: dict[h.group_id].append(h) or dict,
self._homologs.values(), defaultdict(list))
def _db(self, taxid=None):
""" Return an open sqlite3.Connection object.
"""
taxid = taxid or self.taxid
filename = orngServerFiles.localpath_download("PPI",
self.SERVER_FILE.format(taxid=taxid))
if not os.path.exists(filename):
raise ValueError("Database is missing.")
return sqlite3.connect(filename)
def _db(self, taxid=None):
""" Return an open sqlite3.Connection object.
"""
taxid = taxid or self.taxid
filename = orngServerFiles.localpath_download(
"PPI", self.SERVER_FILE.format(taxid=taxid))
if not os.path.exists(filename):
raise ValueError("Database is missing.")
return sqlite3.connect(filename)
def __loadOntologyFromDisk(self):
"""
Function loads MeSH ontology and chemical annotation into internal data structures.
"""
self.toID = dict() # name -> [IDs] Be careful !!! One name can match many IDs!
self.toName = dict() # ID -> name
self.toDesc = dict() # name -> description
self.fromCID = dict() # cid -> term id
self.fromPMID = dict() # pmid -> term id
d = file(orngServerFiles.localpath_download('MeSH', 'mesh-ontology.dat'))
f = file(orngServerFiles.localpath_download('MeSH', 'cid-annotation.dat'))
# loading ontology graph
t = 0
for i in d:
t += 1
parts = i.split("\t") # delimiters are tabs
if(len(parts) != 3):
print "error reading ontology ", parts[0]
parts[2] = parts[2].rstrip("\n\r")
ids = parts[1].split(";")
self.toID[parts[0]] = ids # append additional ID
self.toDesc[parts[0]] = parts[2]
for r in ids:
self.toName[r] = parts[0]
# loading cid -> mesh
for i in f:
parts = i.split(";") # delimiters are tabs
if(len(parts) != 2):
print "error reading ontology ", parts[0]
parts[1] = parts[1].rstrip("\n\r")
cid = int(parts[0])
if self.fromCID.has_key(cid):
self.fromCID[cid].append(parts[1])
else:
self.fromCID[cid] = [parts[1]]
# loading pmid -> mesh, TODO
# print "Current MeSH ontology contains ", t, " mesh terms."
return True
def history(self):
if getattr(self, "_history", None) is None:
fname = orngServerFiles.localpath_download(
"NCBI_geneinfo", "gene_history.%s.db" % self.taxid)
try:
self._history = dict([
(line.split("\t")[2], GeneHistory(line))
for line in open(fname, "rb").read().splitlines()
])
except Exception, ex:
print >> sys.srderr, "Loading NCBI gene history failed.", ex
self._history = {}
def load(self):
import cPickle
dir = orngServerFiles.localpath("EnsembleGeneInfo")
if not os.path.exists(dir):
os.makedirs(dir)
try:
filename = orngServerFiles.localpath_download("EnsembleGeneInfo", self.filename())
info = cPickle.load(open(filename, "rb"))
except Exception, ex:
filename = orngServerFiles.localpath("EnsembleGeneInfo", self.filename())
info = self.create_info()
cPickle.dump(info, open(filename, "wb"))
def load_miRNA_microCosm(org="mus_musculus", max_pvalue=None, min_score=None):
""" Load miRNA's from microcosm into the global scope (currently
only Mus musculus is supported)
"""
global IDs, LABELS, miRNA_lib, mat_toPre, ACCtoID
global preIDs, premiRNA_lib, preACCtoID, clusters
global num_toClusters, clusters_toNum
file = osf.localpath_download("miRNA", "v5.txt.{org}".format(org=org))
[IDs, LABELS, miRNA_lib, mat_toPre, ACCtoID] = parse_targets_microcosm_v5(file,
max_pvalue=max_pvalue, min_score=min_score)
[preIDs, premiRNA_lib, preACCtoID, clusters] = [], {}, {}, {}
num_toClusters, clusters_toNum = {}, {}
def load(self):
import cPickle
dir = orngServerFiles.localpath("EnsembleGeneInfo")
if not os.path.exists(dir):
os.makedirs(dir)
try:
filename = orngServerFiles.localpath_download(
"EnsembleGeneInfo", self.filename())
info = cPickle.load(open(filename, "rb"))
except Exception, ex:
filename = orngServerFiles.localpath("EnsembleGeneInfo",
self.filename())
info = self.create_info()
cPickle.dump(info, open(filename, "wb"))
def load_miRNA_microCosm(org="mus_musculus", max_pvalue=None, min_score=None):
""" Load miRNA's from microcosm into the global scope (currently
only Mus musculus is supported)
"""
global IDs, LABELS, miRNA_lib, mat_toPre, ACCtoID
global preIDs, premiRNA_lib, preACCtoID, clusters
global num_toClusters, clusters_toNum
file = osf.localpath_download("miRNA", "v5.txt.{org}".format(org=org))
[IDs, LABELS, miRNA_lib, mat_toPre,
ACCtoID] = parse_targets_microcosm_v5(file,
max_pvalue=max_pvalue,
min_score=min_score)
[preIDs, premiRNA_lib, preACCtoID, clusters] = [], {}, {}, {}
num_toClusters, clusters_toNum = {}, {}
def __init__(self, organism, genematcher=None):
""" An dictionary like object for accessing NCBI gene info
Arguments::
- *organism* Organism id
Example::
>>> info = NCBIGeneInfo("H**o sapiens")
"""
self.taxid = self.organism_name_search(organism)
fname = orngServerFiles.localpath_download(
"NCBI_geneinfo", "gene_info.%s.db" % self.taxid)
file = open(fname, "rb")
self.update(
dict([(line.split("\t", 3)[1], line)
for line in file.read().splitlines()
if line.strip() and not line.startswith("#")]))
# NOTE orig init time for gene matcher: 2.5s, new 4s: investigate the slowdown
# NOTE matches are not the same because aliases are build a bit
# differently (main name versus old aliases conflict!)
self.matcher = genematcher
if self.matcher == None:
if self.taxid == '352472':
self.matcher = matcher([
GMNCBI(self.taxid),
GMDicty(), [GMNCBI(self.taxid),
GMDicty()]
])
else:
self.matcher = matcher([GMNCBI(self.taxid)])
#if this is done with a gene matcher, pool target names
self.matcher.set_targets(self.keys())
def list_serverfiles():
fname = orngServerFiles.localpath_download(sfdomain, "index.pck")
flist = pickle.load(open(fname, 'r'))
return list_serverfiles_from_flist(flist)
from __future__ import absolute_import, division
from collections import defaultdict
import math, os, random, re, urllib
from Orange.orng import orngServerFiles as osf
import statc
from . import gene as ge, go, kegg as kg, utils, taxonomy as obiTaxonomy
op = utils.stats
mirnafile = osf.localpath_download('miRNA', 'miRNA.txt')
premirnafile = osf.localpath_download('miRNA', 'premiRNA.txt')
################################################################################################################
################################################################################################################
def __build_lib(filename,
labels=True,
MATtoPRE=True,
ACCtoID=True,
clust=False):
"""
build_lib() function takes as input a filename
and gives as output some variables there will be used in
the module.
"""
content = [l.rstrip() for l in open(filename).readlines()][1:]
to_return = []
def create_aliases_version(self):
orngServerFiles.localpath_download("Affy", self.organism + ".pickle")
return orngServerFiles.info("Affy",
self.organism + ".pickle")["datetime"]
def create_aliases(self):
filename = orngServerFiles.localpath_download(
"Affy", self.organism + ".pickle")
import cPickle
return cPickle.load(open(filename, "rb"))
def organism_version(cls, name):
oname = cls.organism_name_search(name)
#FIXME, dirty hack to ensure file id downloaded
orngServerFiles.localpath_download("NCBI_geneinfo", "gene_info.%s.db" % oname)
return orngServerFiles.info("NCBI_geneinfo", "gene_info.%s.db" % oname)["datetime"]
def create_aliases_version(self):
orngServerFiles.localpath_download("Affy", self.organism + ".pickle")
return orngServerFiles.info("Affy", self.organism + ".pickle")["datetime"]
def create_aliases(self):
filename = orngServerFiles.localpath_download("Affy", self.organism + ".pickle")
import cPickle
return cPickle.load(open(filename, "rb"))
def load_serverfiles(hierarchy, organism):
return load_fn(hierarchy, organism, list_serverfiles,
lambda h,o: orngServerFiles.localpath_download(sfdomain, filename(h, o)))
def load(self):
path = orngServerFiles.localpath_download(self.DOMAIN, self.FILENAME)
lines = open(path, "rb").read().splitlines()[:-1]
self._homologs = {}
self._homologs = dict([((h.taxonomy_id, h.gene_symbol), h) for h in [_homolog(line) for line in lines]])
self._homologs_by_group = reduce(lambda dict, h: dict[h.group_id].append(h) or dict, self._homologs.values(), defaultdict(list))
def list_serverfiles():
fname = orngServerFiles.localpath_download(sfdomain, "index.pck")
flist = pickle.load(open(fname, 'r'))
return list_serverfiles_from_flist(flist)
from __future__ import absolute_import, division
from collections import defaultdict
import math, os, random, re, urllib
from Orange.orng import orngServerFiles as osf
import statc
from . import gene as ge, go, kegg as kg, utils, taxonomy as obiTaxonomy
op = utils.stats
mirnafile = osf.localpath_download('miRNA','miRNA.txt')
premirnafile = osf.localpath_download('miRNA','premiRNA.txt')
################################################################################################################
################################################################################################################
def __build_lib(filename, labels=True, MATtoPRE=True, ACCtoID=True, clust=False):
"""
build_lib() function takes as input a filename
and gives as output some variables there will be used in
the module.
"""
content = [l.rstrip() for l in open(filename).readlines()][1:]
to_return = []
ids = [l.split('\t')[0] for l in content]
to_return.append(ids)
if labels:
def __init__(self):
import sqlite3
self.con = sqlite3.connect(
orngServerFiles.localpath_download("HomoloGene",
"InParanoid.sqlite"))
def load_serverfiles(hierarchy, organism):
return load_fn(hierarchy, organism, list_serverfiles,
lambda h,o: orngServerFiles.localpath_download(sfdomain, filename(h, o)))
def __init__(self):
import sqlite3
self.con = sqlite3.connect(orngServerFiles.localpath_download("HomoloGene", "InParanoid.sqlite"))
