def write_genepred_to_fasta_directionless(gpd_filename,ref_fasta,out_fasta):
ofile = open(out_fasta,'w')
ref = sequence_basics.read_fasta_into_hash(ref_fasta)
with open(gpd_filename) as f:
for line in f:
if re.match('^#',line): continue
d = genepred_line_to_dictionary(line)
if d['chrom'] in ref:
seq = ''
for i in range(0,d['exonCount']):
seq = seq+ref[d['chrom']][d['exonStarts'][i]:d['exonEnds'][i]]
ofile.write(">"+str(d['name'])+"\n"+seq+"\n")
ofile.close()
def get_splice_bases(junctions, genome_filename):
genome = sequence_basics.read_fasta_into_hash(genome_filename)
bases = {}
for id in junctions:
chr1 = junctions[id]["chr1"]
coo1 = int(junctions[id]["coo1"])
dir1 = junctions[id]["dir1"]
chr2 = junctions[id]["chr2"]
coo2 = int(junctions[id]["coo2"])
dir2 = junctions[id]["dir2"]
bases1 = "??"
if dir1 == "+" and len(genome[chr1]) > coo1 + 1:
bases1 = genome[chr1][coo1] + genome[chr1][coo1 + 1]
elif dir1 == "-" and len(genome[chr1]) > coo1 - 2:
bases1 = sequence_basics.rc(genome[chr1][coo1 - 3] + genome[chr1][coo1 - 2])
bases2 = "??"
if dir2 == "+" and len(genome[chr2]) > coo2 - 2:
bases2 = genome[chr2][coo2 - 3] + genome[chr2][coo2 - 2]
elif dir2 == "-" and len(genome[chr2]) > coo2 + 1:
bases2 = sequence_basics.rc(genome[chr2][coo2] + genome[chr2][coo2 + 1])
bases[id] = bases1.upper() + " " + bases2.upper()
return bases
def get_splice_bases(junctions,genome_filename):
genome = sequence_basics.read_fasta_into_hash(genome_filename)
bases = {}
for id in junctions:
chr1 = junctions[id]['chr1']
coo1 = int(junctions[id]['coo1'])
dir1 = junctions[id]['dir1']
chr2 = junctions[id]['chr2']
coo2 = int(junctions[id]['coo2'])
dir2 = junctions[id]['dir2']
bases1 = '??'
if dir1 == '+' and len(genome[chr1]) > coo1+1:
bases1 = genome[chr1][coo1] + genome[chr1][coo1+1]
elif dir1 == '-' and len(genome[chr1]) > coo1-2:
bases1 = sequence_basics.rc(genome[chr1][coo1-3] + genome[chr1][coo1-2])
bases2 = '??'
if dir2 == '+' and len(genome[chr2]) > coo2-2:
bases2 = genome[chr2][coo2-3] + genome[chr2][coo2-2]
elif dir2 == '-' and len(genome[chr2]) > coo2+1:
bases2 = sequence_basics.rc(genome[chr2][coo2] + genome[chr2][coo2+1])
bases[id] = bases1.upper() + " " + bases2.upper()
return bases