def gene2exon_list (cursor, gene_id, db_name=None, verbose=False):
exons = []
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select * from gene2exon where gene_id = %d " % gene_id
rows = search_db(cursor, qry)
if (not rows):
rows = search_db(cursor, 'select database()')
if verbose:
print "database ", rows[0][0]
rows = search_db(cursor, qry, verbose = True)
print rows
return []
for row in rows:
exon = Exon()
if (not exon.load_from_gene2exon(row)):
continue
exons.append(exon)
return exons
def gene2exon_list(cursor, gene_id, db_name=None, verbose=False):
exons = []
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select * from gene2exon where gene_id = %d " % gene_id
rows = search_db(cursor, qry)
if not rows:
rows = search_db(cursor, "select database()")
if verbose:
print "database ", rows[0][0]
rows = search_db(cursor, qry, verbose=True)
print rows
return []
for row in rows:
exon = Exon()
if not exon.load_from_gene2exon(row):
continue
exons.append(exon)
return exons
def get_predicted_exons(cursor, gene_id, species):
exons = []
# get the region on the gene
ret = get_gene_region(cursor, gene_id)
if ret:
[gene_seq_id, gene_region_start, gene_region_end, gene_region_strand] = ret
else:
print "region not retrived for ", species, gene_id
return []
qry = "SELECT * FROM prediction_exon WHERE seq_region_id = %d " % gene_seq_id
qry += " AND seq_region_start >= %d AND seq_region_start <= %d " % (gene_region_start, gene_region_end)
qry += " AND seq_region_end >= %d AND seq_region_end <= %d " % (gene_region_start, gene_region_end)
rows = search_db(cursor, qry)
if not rows:
return []
for row in rows:
exon = Exon()
exon.gene_id = gene_id
exon.load_from_ensembl_prediction(gene_region_start, gene_region_end, row)
exons.append(exon)
return exons
def get_predicted_exons (cursor, gene_id, species):
exons = []
# get the region on the gene
ret = get_gene_region (cursor, gene_id)
if ret:
[gene_seq_id, gene_region_start, gene_region_end,
gene_region_strand] = ret
else:
print "region not retrived for ", species, gene_id
return []
qry = "SELECT * FROM prediction_exon WHERE seq_region_id = %d " % gene_seq_id
qry += " AND seq_region_start >= %d AND seq_region_start <= %d " % \
(gene_region_start, gene_region_end)
qry += " AND seq_region_end >= %d AND seq_region_end <= %d " % \
(gene_region_start, gene_region_end)
rows = search_db (cursor, qry)
if (not rows):
return []
for row in rows:
exon = Exon()
exon.gene_id = gene_id
exon.load_from_ensembl_prediction (gene_region_start, gene_region_end, row)
exons.append(exon)
return exons
def get_exon (cursor, exon_id, is_known=None, db_name=None):
exon = Exon ()
if (db_name):
if not switch_to_db(cursor, db_name):
return exon
if is_known==2:
# sw# exon
qry = "select * from sw_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if (not rows):
return exon
exon.load_from_novel_exon (rows[0], "sw_exon")
elif is_known==3:
# sw# exon
qry = "select * from usearch_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if (not rows):
return exon
exon.load_from_novel_exon (rows[0], "usearch_exon")
else:
qry = "select * from gene2exon where exon_id = %d" % exon_id
if is_known: qry += " and is_known = %s " % is_known
rows = search_db(cursor, qry, verbose=False)
if (not rows):
return exon
exon.load_from_gene2exon (rows[0])
return exon
def get_novel_exons (cursor, gene_id, table):
exons = []
qry = "select * from %s " % table
qry += " where gene_id = %d " % int(gene_id)
rows = search_db (cursor, qry)
if not rows: return exons
for row in rows:
exon = Exon()
exon.load_from_novel_exon (row, table)
exons.append(exon)
return exons
def get_exon(cursor, exon_id, is_known=None, db_name=None):
exon = Exon()
if db_name:
if not switch_to_db(cursor, db_name):
return exon
if is_known == 2:
# sw# exon
qry = "select * from sw_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if not rows:
return exon
exon.load_from_novel_exon(rows[0], "sw_exon")
elif is_known == 3:
# sw# exon
qry = "select * from usearch_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if not rows:
return exon
exon.load_from_novel_exon(rows[0], "usearch_exon")
else:
qry = "select * from gene2exon where exon_id = %d" % exon_id
if is_known:
qry += " and is_known = %s " % is_known
rows = search_db(cursor, qry, verbose=False)
if not rows:
return exon
exon.load_from_gene2exon(rows[0])
return exon
def get_known_exons (cursor, gene_id, species):
exons = []
qry = "select distinct exon_transcript.exon_id from exon_transcript, transcript "
qry += " where exon_transcript.transcript_id = transcript.transcript_id "
qry += " and transcript.gene_id = %d " % gene_id
rows = search_db (cursor, qry)
if (not rows ):
return []
if ('Error' in rows[0]):
search_db (cursor, qry, verbose = True)
return []
# get the region on the gene
ret = get_gene_region (cursor, gene_id)
if ret:
[gene_seq_id, gene_region_start, gene_region_end,
gene_region_strand] = ret
else:
print "region not retrived for ", species, gene_id
return []
exon_ids = []
for row in rows:
exon_ids.append(row[0])
for exon_id in exon_ids:
qry = "select * from exon where exon_id=%d" % exon_id
rows = search_db (cursor, qry)
if (not rows or 'Error' in rows[0]):
search_db (cursor, qry, verbose = True)
continue
exon = Exon()
exon.gene_id = gene_id
exon.load_from_ensembl_exon (gene_region_start, gene_region_end, rows[0])
exons.append(exon)
return exons
def extract_regions_from_annotations(self, annotation_lines, region_name):
''' gets the annotation file lines
returns a dictionary of exons positions, keys are the start position
value is a list [start_pos, end_pos, chromosome,
strand(- or + ), and the name of the gene]
also returns the sorted keys of this dictionary (sorted start positions)
'''
exons = ExonsList()
for idx, line in enumerate(annotation_lines):
line = line.split('\t')
if line[2] == region_name:
exon = Exon(line)
exons.add_exon(exon)
self.print_progres(idx, len(annotation_lines))
sys.stderr.write("\n")
return exons
def add_exon(self, chr_map, sequence=''):
exon = Exon(self, chr_map, sequence)
self.exons.append(exon)
return
def map2exon(cursor, ensembl_db_name, map, paralogue=False):
# this is fake exon info! to be passe to get_exon_pepseq
exon = Exon ()
exon.exon_id = map.exon_id_2
exon.is_known = map.exon_known_2
if map.source == 'sw_sharp':
exon.analysis_id = -1
if not paralogue: # move to the other species
rows = switch_to_db (cursor, ensembl_db_name[map.species_2])
if not rows:
exon.exon_seq_id = -1
return exon
else:
qry = "select exon_seq_id from sw_exon where exon_id = %d " % exon.exon_id
rows = search_db (cursor, qry)
if not rows or not rows[0][0]:
exon.exon_seq_id = -1
else:
exon.exon_seq_id = int(rows[0][0])
elif map.source == 'usearch':
exon.analysis_id = -2
if not paralogue:
rows = switch_to_db (cursor, ensembl_db_name[map.species_2])
if not rows:
exon.exon_seq_id = -1
return exon
else:
qry = "select exon_seq_id from usearch_exon where exon_id = %d " % exon.exon_id
rows = search_db (cursor, qry)
if not rows or not rows[0][0]:
exon.exon_seq_id = -1
else:
exon.exon_seq_id = int(rows[0][0])
else:
exon.analysis_id = 1
return exon