def test_pattern(self):
paths = bin_paths_from_fasta(self.fasta, pattern_only=True)
self.assertEquals(paths, 'a.%s.*.bin')
paths = bin_paths_from_fasta(self.fasta,
pattern_only=True,
out_dir="/a")
self.assertEquals(paths, '/a/a.%s.*.bin')
def test_paths(self):
paths = bin_paths_from_fasta(self.fasta)
self.assertEquals(paths, ['a.%s.c.bin', 'a.%s.t.bin', 'a.%s.methyltype.bin'])
def test_pattern(self):
paths = bin_paths_from_fasta(self.fasta, pattern_only=True)
self.assertEquals(paths, 'a.%s.*.bin')
paths = bin_paths_from_fasta(self.fasta, pattern_only=True, out_dir="/a")
self.assertEquals(paths, '/a/a.%s.*.bin')
def test_paths(self):
paths = bin_paths_from_fasta(self.fasta)
self.assertEquals(paths,
['a.%s.c.bin', 'a.%s.t.bin', 'a.%s.methyltype.bin'])
def parse_gsnap_sam(gsnap_f, ref_path, out_dir, paired_end):
fa = Fasta(ref_path)
fc, ft, fmethyltype = \
bin_paths_from_fasta(fa.fasta_name, out_dir)
counts = get_counts(fc, ft, fa)
chr_lengths = dict((k, len(fa[k])) for k in fa.iterkeys())
print >>sys.stderr, "tabulating methylation"
gsnap_subset = open(gsnap_f.replace(".gsnap.sam", ".sam"), "w")
for sline in open(gsnap_f):
if sline.startswith("@SQ"):
print >>gsnap_subset, sline.strip()
continue
# the ends didn't map to same spot.
line = sline.split("\t")
sam_flag = int(line[1])
if paired_end:
if line[6] != "=": continue
#print >>gsnap_subset, sline.strip()
else:
# no reported alignments.
if sam_flag == 4: continue
print >>gsnap_subset, sline.rstrip("\n")
seqid = line[2]
aln_seq = line[9]
read_length = len(aln_seq)
bp0 = int(line[3]) - 1
ga = ((sam_flag & 16) != 0) ^ (sam_flag & 128 != 0)
insert_length = int(line[8])
#line[9] = aln_seq
#line[10] = line[10][:len(aln_seq)]
# both ends start at exactly the same place.
if insert_length == 0: continue
# handle overlapping reads. one side has + insert, the other is -
if -read_length < insert_length < 0:
insert_length = abs(insert_length)
aln_seq = aln_seq[:-(read_length - insert_length)]
read_length = len(aln_seq)
if line[7] == '0': continue
bp1 = bp0 + read_length
ref_seq = (fa[seqid][bp0:bp1]).upper()
letters = 'GA' if ga else 'CT'
read_length = len(ref_seq)
assert read_length > 0, (bp0, bp1)
_update_conversions(ref_seq, aln_seq, bp0, letters,
counts[seqid]['c'], counts[seqid]['t'],
50, read_length, line[5])
write_files(fa.fasta_name, out_dir, counts)
cmd = open(out_dir +"/cmd.ran", "w")
import datetime
print >>cmd, "#date:", str(datetime.date.today())
print >>cmd, "#path:", op.abspath(".")
print >>cmd, " ".join(sys.argv)
write_sam_commands(out_dir, fa, "methylcoder.gsnap")