def splicejunctions_to_gff3(inputBedFile, chrom_sizes, output): gff3 = open(output, 'w') gff3.write("##gff-version 3\n") sizes_dict = subtools.sequence_region(chrom_sizes) seq_regions = dict() with open(inputBedFile, 'r') as bed: for line in bed: field = OrderedDict() attribute = OrderedDict() li = line.rstrip().split("\t") field['seqid'] = li[0] if field['seqid'] not in seq_regions: end_region = sizes_dict[field['seqid']] gff3.write("##sequence-region " + field['seqid'] + ' 1 ' + str(end_region) + '\n') seq_regions[field['seqid']] = end_region field['source'] = li[3] field['type'] = 'junction' # The first base in a chromosome is numbered 0 in BED format field['start'] = int(li[1]) + 1 field['end'] = li[2] field['score'] = li[12] field['strand'] = li[5] field['phase'] = '.' attribute['ID'] = li[0] + '_' + li[3] attribute['Name'] = li[3] attribute['blockcount'] = li[9] attribute['blocksizes'] = li[10] attribute['chromstarts'] = li[11] subtools.write_features(field, attribute, gff3) subtools.child_blocks(field, attribute, gff3, 'exon_junction') gff3.close()
def gtfToGff3(gtf_file, gff3_file, chrom_sizes): """ Covert gtf file output from StringTie to gff3 format """ gff3 = open(gff3_file, 'w') gff3.write("##gff-version 3\n") sizes_dict = subtools.sequence_region(chrom_sizes) seq_regions = dict() parents = dict() with open(gtf_file, 'r') as gtf: for line in gtf: if line.startswith('#') or not line.strip(): continue field = OrderedDict() attribute = OrderedDict() li = line.rstrip().split("\t") #print li field['seqid'] = li[0] #print field['seqid'] if field['seqid'] not in seq_regions: end_region = sizes_dict[field['seqid']] gff3.write("##sequence-region " + field['seqid'] + ' 1 ' + str(end_region) + '\n') seq_regions[field['seqid']] = end_region field['source'] = li[1] field['type'] = li[2] # The first base in a chromosome is numbered 0 in BED format field['start'] = li[3] field['end'] = li[4] field['score'] = li[5] field['strand'] = li[6] field['phase'] = li[7] attr_li = li[8].split(';') gene_id = attr_li[0].split()[1].strip('"') attribute['ID'] = gene_id + '_' + field['type'] + '_' + str( field['start']) + '_' + str(field['end']) if field['type'] == 'transcript': parents[gene_id] = attribute['ID'] attribute['transcript_id'] = attr_li[1].split()[1].strip('"') attribute['coverage'] = attr_li[2].split()[1].strip('"') attribute['fpkm'] = attr_li[3].split()[1].strip('"') attribute['tpm'] = attr_li[4].split()[1].strip('"') elif field['type'] == 'exon': attribute['Parent'] = parents[gene_id] attribute['transcript_id'] = attr_li[1].split()[1].strip('"') attribute['coverage'] = attr_li[3].split()[1].strip('"') subtools.write_features(field, attribute, gff3) gff3.close()
def bigpsl_to_gff3(self): gff3 = open(self.gff3_file.name, 'w') gff3.write("##gff-version 3\n") sizes_dict = subtools.sequence_region(self.chromSizesFile) seq_regions = dict() with open(self.inputFile, 'r') as bed: for line in bed: field = OrderedDict() attribute = OrderedDict() li = line.rstrip().split("\t") field['seqid'] = li[0] if field['seqid'] not in seq_regions: end_region = sizes_dict[field['seqid']] gff3.write("##sequence-region " + field['seqid'] + ' 1 ' + str(end_region) + '\n') seq_regions[field['seqid']] = end_region field['source'] = 'UCSC BLAT alignment tool' field['type'] = 'match' # The first base in a chromosome is numbered 0 in BED format field['start'] = str(int(li[1]) + 1) field['end'] = li[2] field['score'] = li[4] field['strand'] = li[5] field['phase'] = '.' attribute['ID'] = li[0] + '_' + li[3] attribute['Name'] = li[3] attribute['blockcount'] = li[9] attribute['blocksizes'] = li[10] attribute['chromstarts'] = li[11] attribute['ochrom_start'] = li[12] attribute['ochrom_end'] = li[13] attribute['ochrom_strand'] = li[14] attribute['ochrom_size'] = li[15] attribute['ochrom_starts'] = li[16] attribute['sequence on other chromosome'] = li[17] attribute['cds in ncbi format'] = li[18] attribute['size of target chromosome'] = li[19] attribute['number of bases matched'] = li[20] attribute['number of bases that don\'t match'] = li[21] attribute[ 'number of bases that match but are part of repeats'] = li[ 22] attribute['number of \'N\' bases'] = li[23] subtools.write_features(field, attribute, gff3) subtools.child_blocks(field, attribute, gff3, 'match_part') gff3.close()
def trfbig_to_gff3(inputBedFile, chrom_sizes, output): gff3 = open(output, 'w') gff3.write("##gff-version 3\n") sizes_dict = subtools.sequence_region(chrom_sizes) seq_regions = dict() with open(inputBedFile, 'r') as bed: for line in bed: field = OrderedDict() attribute = OrderedDict() li = line.rstrip().split("\t") field['seqid'] = li[0] if field['seqid'] not in seq_regions: end_region = sizes_dict[field['seqid']] gff3.write("##sequence-region " + field['seqid'] + ' 1 ' + str(end_region) + '\n') seq_regions[field['seqid']] = end_region field['source'] = li[3] field['type'] = 'tandem_repeat' # The first base in a chromosome is numbered 0 in BED format field['start'] = str(int(li[1]) + 1) field['end'] = li[2] field['score'] = li[9] field['strand'] = '+' field['phase'] = '.' attribute['length of repeat unit'] = li[4] attribute['mean number of copies of repeat'] = li[5] attribute['length of consensus sequence'] = li[6] attribute['percentage match'] = li[7] attribute['percentage indel'] = li[8] attribute['percent of a\'s in repeat unit'] = li[10] attribute['percent of c\'s in repeat unit'] = li[11] attribute['percent of g\'s in repeat unit'] = li[12] attribute['percent of t\'s in repeat unit'] = li[13] attribute['entropy'] = li[14] attribute['sequence of repeat unit element'] = li[15] subtools.write_features(field, attribute, gff3) gff3.close()