def load_mouse_ortho_data(
filename = os.path.join(biopsy.get_data_dir(), 'TreeFam', 'orthologs', 'MOUSE_ORTHO.tsv')
):
reader = csv.reader(
open( filename, "rb" ),
delimiter = '\t'
)
return one_to_many( generate_db_refs_for_mouse_ortho( reader ) )
def to_mouse_map(ortho_filename = os.path.join(biopsy.get_data_dir(), 'TreeFam', 'orthologs', 'MOUSE_ORTHO.tsv')):
return dict(
(biopsy.DbRef.try_to_parse(row[1]), biopsy.DbRef.try_to_parse(row[0]))
for row
in csv.reader(
open(ortho_filename),
delimiter = '\t'
)
if len(row) > 1
)
mapper[T.db.entrez_gene] = entrez_gene_mapper
mapper[T.db.swissprot] = uniprot_mapper
return map, mapper
def build_map():
map, mapper = default_map_and_mapper()
for m in T.Matrix.all():
mapper(m.acc.as_db_ref())
for s in T.Site.all():
if s.id.factor != 'CONS':
continue
mapper(s.acc.as_db_ref())
return map
identifier_map = PersistedCache(build_map, os.path.join(biopsy.get_data_dir(), 'identifiers', 'identifier_map.pickle'))
def small_test_map():
'Return a small map for testing purposes.'
map, mapper = default_map_and_mapper()
mapper(T.DbRef.parse_as('71431', T.db.entrez_gene))
return map
def matrices_that_map_to(map, db):
"""
Return a set of those matrices that map to the at least one entry in the
given database type
"""
matrices = set()
for m in T.Matrix.all():
mouse_genes.add(r)
return mouse_genes
def write_ensembl_mouse_genes_file(filename):
"""
Writes all the ensembl mouse gene references in transfac gene table to a
file for use at
http://www.informatics.jax.org/javawi2/servlet/WIFetch?page=batchQF.
"""
f = open(filename, "w")
for g in ensembl_mouse_genes_in_transfac():
f.write(str(g))
f.write('\n')
f.close()
accession_map_filename = os.path.join(biopsy.get_data_dir(), 'identifiers', 'mgi', 'MRK_Dump1.rpt')
def parse_accession_map():
"""
Parse the MGI accession map flat file and yield tuples
(mgi identifier ref, mgi accession)
"""
for id, acc in csv.reader(open(accession_map_filename, 'rb'), delimiter='\t'):
yield (T.DbRef.parse_as(id, T.db.mgi), acc)
def build_acc_2_id_map():
"""
Returns a dict mapping MGI accessions to ids.
"""
return dict((acc, id) for id, acc in parse_accession_map())
def get_kegg_pathways():
pathway_dir = os.path.join(biopsy.get_data_dir(), 'KEGG', 'pathways')
return [
(file, set(l.strip() for l in open(os.path.join(pathway_dir, file))))
for file in os.listdir(pathway_dir)
]
def get_data_dir():
return os.path.join(biopsy.get_data_dir(), 'SymAtlas')
Query Entrez to get a map from its protein accessions to ids and xrefs
"""
result = ProteinMap(
acc_2_id = cookbook.DictOfSets(),
xrefs = cookbook.DictOfSets()
)
for acc, id, refs in refs_for_mouse_protein_accs():
result.acc_2_id[acc].add(id.acc)
for ref in refs:
result.xrefs[id.acc].add(ref)
return result
_proteins_pickle_file = os.path.join(biopsy.get_data_dir(), 'identifiers', 'entrez', 'proteins.pickle')
proteins = lazy.PersistableLazyInitialiser(get_protein_map, _proteins_pickle_file)
def write_mouse_protein_ids(filename):
from Bio.EUtils import HistoryClient
f = open(filename, 'w')
results = HistoryClient.HistoryClient().search(db='protein', term='mouse[orgn]')
for id in results.dbids.ids:
f.write(id)
f.write('\n')
f.close()
self.marginal_prior,
self.joint_prior,
self.mi_threshold
)
def sequences_from_jaspar_file(jaspar_file):
length = None
for l in jaspar_file:
if l.startswith('>'): continue
if not l.strip(): continue
site = ''.join( [ c for c in l if c.isupper() ] )
if length == None: length = len(site)
elif len(site) != length: continue
yield site
_jaspar_dir = os.path.join(biopsy.get_data_dir(), 'Jaspar', 'JASPAR_CORE')
_jaspar_phylofacts_dir = os.path.join(biopsy.get_data_dir(), 'Jaspar', 'JASPAR_PHYLOFACTS')
def jaspar_sequences(dir = _jaspar_dir):
import os
for filename in os.listdir(dir):
f = open(os.path.join(dir, filename), 'r')
yield (
filename.split('.')[0],
[ s for s in sequences_from_jaspar_file(f) ]
)
def jaspar_phylofacts_sequences():
for x in jaspar_sequences(_jaspar_phylofacts_dir):
yield x
if '__main__' == __name__:
def get_data_dir():
return os.path.join(biopsy.get_data_dir(), 'biobase', 'transpro')
def get_programs_dir():
"@return: The directory to put program specific info into."
#ensure_dir_exists(_programs_dir)
return _programs_dir
_summaries_dir = os.path.join(options.output_dir, 'summaries')
"the directory to put DPM summaries into"
def get_summaries_dir():
"@return: The directory to put DPM summaries into."
#ensure_dir_exists(_summaries_dir)
return _summaries_dir
_site_dpm_data_dir = os.path.join(biopsy.get_data_dir(), 'site-dpm')
"The directory where results that are reused across runs are cached."
def get_site_dpm_data_dir():
"@return: The directory where results that are reused across runs are cached."
#ensure_dir_exists(_site_dpm_data_dir)
return _site_dpm_data_dir
try:
import pylab
except:
import warnings
warnings.warn('Could not set matplotlib figure size')
print sys.exc_info()
def get_data_dir():
return os.path.join(biopsy.get_data_dir(), 'biobase', 'transpro')
#
# Copyright John Reid 2006-2010
#
import gzip, os.path, re, biopsy
_base_data_dir = os.path.join(biopsy.get_data_dir(), 'ensembl', 'genomes')
def get_genome_dir(genome):
return os.path.join(_base_data_dir, genome)
def get_chromosome_file(genome, chromosome):
"""Returns an open file handle for the chromosome file in the given genome"""
genome_dir = get_genome_dir(genome)
files = os.listdir(genome_dir)
chr_re = re.compile('\.dna\.chromosome\.%s$' % (str(chromosome)))
matched_files = [f for f in files if chr_re.search(f)]
if 0 == len(matched_files):
raise RuntimeError('Did not find chromosome "%s" in directory: %s' %
(str(chromosome), genome_dir))
if len(matched_files) > 1:
raise RuntimeError('Expecting only one match in genome directory: %s' %
genome_dir)
filename = os.path.join(genome_dir, matched_files[0])
print 'File: %s' % filename
return open(filename, 'r')
def get_chromosome_sequence(chromosome_file, offset, length):
# Copyright John Reid 2007
#
"""
Code to parse UniProt data.
www.ensembl.org/
"""
import gzip, sys, cookbook, biopsy
from . import lazy
T = biopsy.transfac
_uniprot_file = os.path.join(biopsy.get_data_dir(), "UniProt", "uniprot_sprot.dat.gz")
def data():
"""
Returns a handle to the uniprot data.
"""
return gzip.open(_uniprot_file)
def yield_records(handle):
"""
Takes a file like handle and separates it into records.
"""
record = []
for line in handle:
#
# Copyright John Reid 2006-2010
#
import gzip, os.path, re, biopsy
_base_data_dir = os.path.join(biopsy.get_data_dir(), 'ensembl', 'genomes')
def get_genome_dir( genome ):
return os.path.join( _base_data_dir, genome )
def get_chromosome_file( genome, chromosome ):
"""Returns an open file handle for the chromosome file in the given genome"""
genome_dir = get_genome_dir( genome )
files = os.listdir( genome_dir )
chr_re = re.compile( '\.dna\.chromosome\.%s$' % ( str( chromosome ) ) )
matched_files = [ f for f in files if chr_re.search( f ) ]
if 0 == len( matched_files ):
raise RuntimeError(
'Did not find chromosome "%s" in directory: %s' % (
str( chromosome ),
genome_dir
)
)
if len( matched_files ) > 1:
raise RuntimeError(
'Expecting only one match in genome directory: %s' % genome_dir
)
filename = os.path.join( genome_dir, matched_files[0] )
print 'File: %s' % filename
return open( filename, 'r' )
