def write_gene_to_gtf(ofp, gene):
lines = []
for index, transcript in enumerate(gene.transcripts):
meta_data = {}
if config.FIX_CHRM_NAMES_FOR_UCSC:
transcript.chrm = fix_chrm_name_for_ucsc(transcript.chrm)
assert transcript.gene_id != None
lines.append(
transcript.build_gtf_lines(meta_data, source="grit") + "\n")
ofp.write("".join(lines))
return
def write_gene_to_gtf( ofp, gene ):
lines = []
for index, transcript in enumerate(gene.transcripts):
meta_data = {}
if config.FIX_CHRM_NAMES_FOR_UCSC:
transcript.chrm = fix_chrm_name_for_ucsc(transcript.chrm)
assert transcript.gene_id != None
lines.append( transcript.build_gtf_lines(
meta_data, source="grit") + "\n" )
ofp.write( "".join(lines) )
return
def writeGff( self, ofp ):
"""
chr7 127471196 127472363 Pos1 0 + 127471196 127472363 255,0,0
"""
if self.strand == '-':
writetable_bins = self.reverse_strand( contig_len )
else:
writetable_bins = self
for bin in writetable_bins:
if filter != None and bin.type != filter:
continue
chrm = elements.chrm
if config.FIX_CHRM_NAMES_FOR_UCSC:
chrm = fix_chrm_name_for_ucsc(self.chrm)
# add 1 because gffs are 1-based
region = GenomicInterval(chrm, self.strand,
bin.start+1, bin.stop+1)
grp_id = "%s_%s_%i_%i" % region
ofp.write( create_gff_line(region, grp_id) + "\n" )
return
def write_gene_to_tracking_file(ofp, gene):
lines = []
contig_name = gene.chrm
if config.FIX_CHRM_NAMES_FOR_UCSC:
contig_name = fix_chrm_name_for_ucsc(contig_name)
for t in gene.transcripts:
if t.gene_name != None:
gene_short_name = t.gene_name
elif t.ref_gene != None:
gene_short_name = t.ref_gene
else:
gene_short_name = '-'
line = [
# tracking ID
(t.id).ljust(20),
# class code
('-' if t.ref_match_class_code == None else t.ref_match_class_code
).ljust(10),
# nearest ref id
('-' if t.ref_trans == None else t.ref_trans).ljust(20),
# gene unique id
(t.gene_id).ljust(20),
# gene short name
('-' if t.gene_name == None else t.gene_name).ljust(20),
# TSS ID
('-').ljust(10),
("%s:%s:%i-%i" %
(contig_name, t.strand, t.start, t.stop)).ljust(30),
# transcript length
str(t.calc_length())
]
lines.append("\t".join(line))
ofp.write("\n".join(lines) + "\n")
return
def write_gene_to_tracking_file( ofp, gene):
lines = []
contig_name = gene.chrm
if config.FIX_CHRM_NAMES_FOR_UCSC:
contig_name = fix_chrm_name_for_ucsc(contig_name)
for t in gene.transcripts:
if t.gene_name != None:
gene_short_name = t.gene_name
elif t.ref_gene != None:
gene_short_name = t.ref_gene
else:
gene_short_name = '-'
line = [
# tracking ID
(t.id).ljust(20),
# class code
('-' if t.ref_match_class_code == None
else t.ref_match_class_code).ljust(10),
# nearest ref id
('-' if t.ref_trans == None else t.ref_trans).ljust(20),
# gene unique id
(t.gene_id).ljust(20),
# gene short name
('-' if t.gene_name == None else t.gene_name).ljust(20),
# TSS ID
('-').ljust(10),
("%s:%s:%i-%i"%(contig_name, t.strand, t.start, t.stop)).ljust(30),
# transcript length
str(t.calc_length()) ]
lines.append("\t".join(line))
ofp.write( "\n".join(lines) + "\n" )
return
def write_elements_bed( self, ofp ):
feature_mapping = {
'GENE': 'gene',
'CAGE_PEAK': 'promoter',
'SE_GENE': 'single_exon_gene',
'TSS_EXON': 'tss_exon',
'EXON': 'internal_exon',
'TES_EXON': 'tes_exon',
'INTRON': 'intron',
'POLYA': 'polya',
'INTERGENIC_SPACE': 'intergenic',
'RETAINED_INTRON': 'retained_intron',
'UNKNOWN': 'UNKNOWN'
}
color_mapping = {
'GENE': '200,200,200',
'CAGE_PEAK': '153,255,000',
'SE_GENE': '000,000,200',
'TSS_EXON': '140,195,59',
'EXON': '000,000,000',
'TES_EXON': '255,51,255',
'INTRON': '100,100,100',
'POLYA': '255,0,0',
'INTERGENIC_SPACE': '254,254,34',
'RETAINED_INTRON': '255,255,153',
'UNKNOWN': '0,0,0'
}
chrm = self.chrm
if config.FIX_CHRM_NAMES_FOR_UCSC:
chrm = fix_chrm_name_for_ucsc(chrm)
# write the gene line
bed_line = create_bed_line( chrm, self.strand,
self.start, self.stop+1,
feature_mapping['GENE'],
score=1000,
color=color_mapping['GENE'],
use_thick_lines=True,
blocks=[(x.start, x.stop) for x in self.regions])
ofp.write( bed_line + "\n" )
try: max_min_fpkm = max(1e-1, max(bin.fpkm for bin in self.elements)) \
if len(self.elements) > 0 else 0
except: max_min_fpkm = 1000
for element in self.elements:
region = ( chrm, self.strand, element.start, element.stop)
blocks = []
use_thick_lines=(element.type != 'INTRON')
element_type = element.type
if element_type == None:
element_type = 'UNKNOWN'
continue
try: fpkm = element.fpkm_ub
except: fpkm = element.fpkm
score = min(1000, int(1000*fpkm/max_min_fpkm))
score = 1000
grp_id = element_type + "_%s_%s_%i_%i" % region
# also, add 1 to stop because beds are open-closed ( which means no net
# change for the stop coordinate )
bed_line = create_bed_line( chrm, self.strand,
element.start, element.stop+1,
feature_mapping[element_type],
score=score,
color=color_mapping[element_type],
use_thick_lines=use_thick_lines,
blocks=blocks)
ofp.write( bed_line + "\n" )
return