def load_ensembl2pfam(self):
#Prosite
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__prot_pfam__dm.txt')
load_start(DATAFILE)
ensembl2pfam = dict_nodup(tab2dict(DATAFILE, (1, 4), 0))
ensembl2pfam = value_convert(ensembl2pfam, lambda x: {'pfam': x}, traverse_list=False)
load_done('[%d]' % len(ensembl2pfam))
return self.convert2entrez(ensembl2pfam)
def load_ensembl2pfam(self):
#Prosite
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__prot_pfam__dm.txt')
load_start(DATAFILE)
ensembl2pfam = dict_nodup(tab2dict(DATAFILE, (1, 4), 0))
ensembl2pfam = value_convert(ensembl2pfam, lambda x: {'pfam': x}, traverse_list=False)
load_done('[%d]' % len(ensembl2pfam))
return self.convert2entrez(ensembl2pfam)
def _load_ensembl_2taxid(self):
"""ensembl2taxid"""
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__translation__main.txt')
load_start(DATAFILE)
ensembl2taxid = dict_nodup(tab2dict(DATAFILE, (0, 1), 1, includefn=_not_LRG))
# need to convert taxid to integer here
ensembl2taxid = value_convert(ensembl2taxid, lambda x: int(x))
load_done('[%d]' % len(ensembl2taxid))
return ensembl2taxid
def load_ensembl2interpro(self):
#Interpro
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__prot_interpro__dm.txt')
load_start(DATAFILE)
ensembl2interpro = dict_nodup(tab2dict(DATAFILE, (1, 4, 5, 6), 0))
ensembl2interpro = value_convert(ensembl2interpro, lambda x: {'id': x[0], 'short_desc': x[1], 'desc': x[2]})
ensembl2interpro = value_convert(ensembl2interpro, lambda x: {'interpro': x}, traverse_list=False)
load_done('[%d]' % len(ensembl2interpro))
return self.convert2entrez(ensembl2interpro)
def load_ensembl2pos(self):
#Genomic position
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__gene__main.txt')
load_start(DATAFILE)
ensembl2pos = dict_nodup(tab2dict(DATAFILE, (1, 3, 4, 5, 6), 0, includefn=_not_LRG))
ensembl2pos = value_convert(ensembl2pos, lambda x: {'chr': x[2], 'start': int(x[0]), 'end': int(x[1]), 'strand': int(x[3])})
ensembl2pos = value_convert(ensembl2pos, lambda x: {'genomic_pos': x}, traverse_list=False)
load_done('[%d]' % len(ensembl2pos))
return self.convert2entrez(ensembl2pos)
def _load_ensembl_2taxid(self):
"""ensembl2taxid"""
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__translation__main.txt')
load_start(DATAFILE)
ensembl2taxid = dict_nodup(tab2dict(DATAFILE, (0, 1), 1, includefn=_not_LRG))
# need to convert taxid to integer here
ensembl2taxid = value_convert(ensembl2taxid, lambda x: int(x))
load_done('[%d]' % len(ensembl2taxid))
return ensembl2taxid
def load_ensembl2interpro(self):
#Interpro
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__prot_interpro__dm.txt')
load_start(DATAFILE)
ensembl2interpro = dict_nodup(tab2dict(DATAFILE, (1, 4, 5, 6), 0))
ensembl2interpro = value_convert(ensembl2interpro, lambda x: {'id': x[0], 'short_desc': x[1], 'desc': x[2]})
ensembl2interpro = value_convert(ensembl2interpro, lambda x: {'interpro': x}, traverse_list=False)
load_done('[%d]' % len(ensembl2interpro))
return self.convert2entrez(ensembl2interpro)
def load_ensembl2pos(self):
#Genomic position
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__gene__main.txt')
load_start(DATAFILE)
ensembl2pos = dict_nodup(tab2dict(DATAFILE, (1, 3, 4, 5, 6), 0, includefn=_not_LRG))
ensembl2pos = value_convert(ensembl2pos, lambda x: {'chr': x[2], 'start': int(x[0]), 'end': int(x[1]), 'strand': int(x[3])})
ensembl2pos = value_convert(ensembl2pos, lambda x: {'genomic_pos': x}, traverse_list=False)
load_done('[%d]' % len(ensembl2pos))
return self.convert2entrez(ensembl2pos)
def convert2entrez(self, ensembl2x):
'''convert a dict with ensembl gene ids as the keys to matching entrezgene ids as the keys.'''
if not self.ensembl2entrez_li:
self._load_ensembl2entrez_li()
if not self.ensembl_main:
self.ensembl_main = self.load_ensembl_main()
ensembl2entrez = list2dict(self.ensembl2entrez_li, 0)
entrez2ensembl = list2dict(self.ensembl2entrez_li, 1)
#Now make a dictionary indexed by entrez gene id
print('# of ensembl IDs in total: %d' %
len(set(ensembl2x) | set(ensembl2entrez)))
print('# of ensembl IDs match entrez Gene IDs: %d' %
len(set(ensembl2x) & set(ensembl2entrez)))
print('# of ensembl IDs DO NOT match entrez Gene IDs: %d' %
len(set(ensembl2x) - set(ensembl2entrez)))
#all genes with matched entrez
def _fn(eid, taxid=None):
d = copy.copy(ensembl2x.get(
eid, {})) # need to make a copy of the value here.
return d # otherwise, it will cause issue when multiple entrezgene ids
# match the same ensembl gene, for example,
# ENSMUSG00000027104 --> (11909, 100047997)
data = value_convert(entrez2ensembl, _fn)
#add those has no matched entrez geneid, using ensembl id as the key
for eid in (set(ensembl2x) - set(ensembl2entrez)):
_g = ensembl2x[eid]
#_g.update(self.ensembl_main.get(eid, {}))
data[eid] = _g
for id in data:
if isinstance(data[id], dict):
_doc = dict_nodup(data[id], sort=True)
else:
#if one entrez gene matches multiple ensembl genes
_doc = dict_attrmerge(data[id], removedup=True, sort=True)
data[id] = _doc
return data
def load_ensembl2pos(self):
#Genomic position
DATAFILE = os.path.join(DATA_FOLDER, 'gene_ensembl__gene__main.txt')
load_start(DATAFILE)
# Twice 1 because first is the dict key, the second because we need gene id within genomic_pos
ensembl2pos = dict_nodup(
tab2dict(DATAFILE, (1, 1, 3, 4, 5, 6), 0, includefn=_not_LRG))
ensembl2pos = value_convert(
ensembl2pos, lambda x: {
'ensemblgene': x[0],
'chr': x[3],
'start': int(x[1]),
'end': int(x[2]),
'strand': int(x[4])
})
ensembl2pos = value_convert(ensembl2pos,
lambda x: {'genomic_pos': x},
traverse_list=False)
load_done('[%d]' % len(ensembl2pos))
return self.convert2entrez(ensembl2pos)
def convert2entrez(self, ensembl2x):
'''convert a dict with ensembl gene ids as the keys to matching entrezgene ids as the keys.'''
if not self.ensembl2entrez_li:
self._load_ensembl2entrez_li()
if not self.ensembl_main:
self.ensembl_main = self.load_ensembl_main()
ensembl2entrez = list2dict(self.ensembl2entrez_li, 0)
entrez2ensembl = list2dict(self.ensembl2entrez_li, 1)
#Now make a dictionary indexed by entrez gene id
print '# of ensembl IDs in total: %d' % len(set(ensembl2x) | set(ensembl2entrez))
print '# of ensembl IDs match entrez Gene IDs: %d' % len(set(ensembl2x) & set(ensembl2entrez))
print '# of ensembl IDs DO NOT match entrez Gene IDs: %d' % len(set(ensembl2x) - set(ensembl2entrez))
#all genes with matched entrez
def _fn(eid, taxid=None):
d = copy.copy(ensembl2x.get(eid, {})) #need to make a copy of the value here.
return d #otherwise, it will cause issue when multiple entrezgene ids
#match the same ensembl gene, for example,
# ENSMUSG00000027104 --> (11909, 100047997)
data = value_convert(entrez2ensembl, _fn)
#add those has no matched entrez geneid, using ensembl id as the key
for eid in (set(ensembl2x) - set(ensembl2entrez)):
_g = ensembl2x[eid]
#_g.update(self.ensembl_main.get(eid, {}))
data[eid] = _g
doc_li = []
for id in data:
if type(data[id]) is types.DictType:
_doc = dict_nodup(data[id], sort=True)
else:
#if one entrez gene matches multiple ensembl genes
_doc = dict_attrmerge(data[id], removedup=True, sort=True)
data[id] = _doc
return data
