def ucsc_tfbs_conserved_parse(self, curs, inputfile, gene_symbol2gene_id, tax_id):
"""
2008-08-12
return_target_gene_ls() returns a different data structure.
03-29-06
--get_entrezgene_annotated_anchor()
--submit2matrix()
--return_target_gene_ls()
--find_binding_site_disp_coord()
--get_sequence_segment()
--submit2binding_site()
"""
sys.stderr.write("Parsing ucsc_tfbs_conserved ...\n")
chromosome2anchor_gene_tuple_ls, gene_id2coord = get_entrezgene_annotated_anchor(curs, tax_id)
inf = open(inputfile, 'r')
mt_id_set = Set()
line_no = 0
for line in inf:
line_no+=1
if line[0]=='#': #skip the comment lines
continue
row = line[:-1].split('\t')
try:
bin, chromosome, start, end, name, score, strand, zscore = row
chromosome = chromosome[3:]
binding_site_attr = binding_site_attribute(tax_id=tax_id)
if chromosome.find('random')==-1: #03-29-06 not random sequences
binding_site_attr.chromosome = chromosome
binding_site_attr.strand = strand
mt_id = name
if mt_id not in mt_id_set:
self.submit2matrix(curs, mt_id, tax_id)
mt_id_set.add(mt_id)
binding_site_attr.mt_id = mt_id
binding_site_attr.bs_genome_start = int(start)
binding_site_attr.bs_genome_end = int(end)
binding_site_attr.matrix_similarity_score = float(score)
binding_site_attr.core_similarity_score = float(zscore)
regulatory_coord = (binding_site_attr.chromosome, \
binding_site_attr.bs_genome_start, binding_site_attr.bs_genome_end)
target_gene_ls, target_gene_ls_type = self.return_target_gene_ls(regulatory_coord, \
chromosome2anchor_gene_tuple_ls, gene_id2coord)
pdata = self.return_target_gene_ls(regulatory_coord, chromosome2anchor_gene_tuple_ls, gene_id2coord)
if pdata.regulatory_touch_target_gene_ls:
target_gene_ls_type = 'touch'
target_gene_ls = pdata.regulatory_touch_target_gene_ls
elif pdata.regulatory_is_left_upstream_target_gene_ls or pdata.regulatory_is_right_upstream_target_gene_ls:
target_gene_ls_type = 'upstream'
target_gene_ls = pdata.regulatory_is_left_upstream_target_gene_ls + pdata.regulatory_is_right_upstream_target_gene_ls
binding_site_attr.comment = 'bin:%s. %s'%(bin, target_gene_ls_type)
for target_gene_tuple in target_gene_ls:
anchor, gene_id = target_gene_tuple[:2]
binding_site_attr.prom_id = gene_id
gene_start, gene_stop, gene_strand, gene_genomic_gi = gene_id2coord[gene_id]
self.find_binding_site_disp_coord(binding_site_attr, gene_strand, gene_start, gene_stop)
binding_site_attr.sequence = get_sequence_segment(curs, gene_genomic_gi, \
binding_site_attr.bs_genome_start, binding_site_attr.bs_genome_end)
if strand == '-': #reverse_complement
seq = Seq(binding_site_attr.sequence)
binding_site_attr.sequence = seq.reverse_complement().tostring()
self.submit2binding_site(curs, binding_site_attr)
except:
print "line_no:", line_no
print line
sys.stderr.write("Done.\n")
def sgd_regulatory_parse(self, curs, inputfile, gene_symbol2gene_id, tax_id):
"""
2008-08-12
return_target_gene_ls() returns a different data structure.
12-17-05
--get_entrezgene_annotated_anchor()
--calculate_rome_number()
--parse_sgd_regulatory_attribute()
--submit2matrix()
--return_target_gene_ls()
--find_binding_site_disp_coord()
--get_sequence_segment()
--submit2binding_site()
"""
sys.stderr.write("Parsing sgd_regulatory ...\n")
chromosome2anchor_gene_tuple_ls, gene_id2coord = get_entrezgene_annotated_anchor(curs, tax_id)
inf = open(inputfile, 'r')
mt_id_set = Set()
line_no = 0
for line in inf:
line_no+=1
if line[0]=='#': #skip the comment lines
continue
row = line[:-1].split('\t')
try:
seqname, source, feature, start, end, score, strand, frame, attribute = row
chromosome_rome_number = seqname[3:]
if feature=='TF_binding_site':
binding_site_attr = binding_site_attribute(tax_id=tax_id)
chromosome = calculate_rome_number(chromosome_rome_number)
if chromosome:
binding_site_attr.chromosome = repr(chromosome) #varchar type in database
else: #it might be something like Mito, return None
binding_site_attr.chromosome = chromosome_rome_number
binding_site_attr.strand = strand
mt_id, dbxref = self.parse_sgd_regulatory_attribute(attribute)
if mt_id not in mt_id_set:
self.submit2matrix(curs, mt_id, tax_id)
mt_id_set.add(mt_id)
binding_site_attr.mt_id = mt_id
binding_site_attr.bs_genome_start = int(start)
binding_site_attr.bs_genome_end = int(end)
regulatory_coord = (binding_site_attr.chromosome, \
binding_site_attr.bs_genome_start, binding_site_attr.bs_genome_end)
pdata = self.return_target_gene_ls(regulatory_coord, chromosome2anchor_gene_tuple_ls, gene_id2coord)
if pdata.regulatory_touch_target_gene_ls:
target_gene_ls_type = 'touch'
target_gene_ls = pdata.regulatory_touch_target_gene_ls
elif pdata.regulatory_is_left_upstream_target_gene_ls or pdata.regulatory_is_right_upstream_target_gene_ls:
target_gene_ls_type = 'upstream'
target_gene_ls = pdata.regulatory_is_left_upstream_target_gene_ls + pdata.regulatory_is_right_upstream_target_gene_ls
binding_site_attr.comment = '%s;%s'%(dbxref, target_gene_ls_type)
for target_gene_tuple in target_gene_ls:
anchor, gene_id = target_gene_tuple[:2]
binding_site_attr.prom_id = gene_id
gene_start, gene_stop, gene_strand, gene_genomic_gi = gene_id2coord[gene_id]
self.find_binding_site_disp_coord(binding_site_attr, gene_strand, gene_start, gene_stop)
binding_site_attr.sequence = get_sequence_segment(curs, gene_genomic_gi, \
binding_site_attr.bs_genome_start, binding_site_attr.bs_genome_end)
if strand == '-': #reverse_complement
seq = Seq(binding_site_attr.sequence)
binding_site_attr.sequence = seq.reverse_complement().tostring()
self.submit2binding_site(curs, binding_site_attr)
except:
print "line_no:", line_no
print line
sys.stderr.write("Done.\n")