def testFromVCF(self):
self.vcf = pysam.VCF()
self.assertRaises(
UnicodeDecodeError,
self.vcf.connect, self.tmpfilename + ".gz", "ascii")
self.vcf.connect(self.tmpfilename + ".gz", encoding="utf-8")
v = self.vcf.getsamples()[0]
def __init__(self, filename, sample):
self.sample = sample
self.vcf = pysam.VCF()
self.vcf.connect(filename)
if sample not in self.vcf.getsamples():
raise KeyError("sample %s not vcf file" % sample)
def load(self, filename, exclude=None):
'''load tracks from a vcf file.'''
tracks = []
v = pysam.VCF()
v.setversion(40)
if not os.path.exists(filename):
self.tracks = tracks
return self
v.connect(filename)
if exclude:
to_exclude = [re.compile(x) for x in exclude]
for sample in v.getsamples():
if exclude:
for x in to_exclude:
if x.search(sample):
skip = True
break
if skip:
continue
tracks.append(self.factory(sample))
self.tracks = tracks
return self
def iterate_from_vcf(infile, sample):
'''iterate over a vcf-formatted file.
*infile* can be any iterator over a lines.
The function yields named tuples of the type
:class:`pysam.Pileup.PileupSubstitution` or
:class:`pysam.Pileup.PileupIndel`.
Positions without a snp will be skipped.
This method is wasteful and written to support same legacy code
that expects samtools pileup output.
Better use the vcf parser directly.
'''
vcf = pysam.VCF()
vcf.connect(infile)
if sample not in vcf.getsamples():
raise KeyError("sample %s not vcf file")
for row in vcf.fetch():
result = vcf2pileup(row, sample)
if result:
yield result
def testRead(self):
self.vcf = pysam.VCF()
self.vcf.connect(self.filename)
for x in self.vcf.fetch():
print str(x)
print x.pos
print x.alt
print x.id
print x.qual
print x.filter
print x.info
print x.format
for s in x.samples:
print s, x[s]
def setUp(self):
TestVCF.setUp(self)
self.vcf = pysam.VCF()
self.compare = loadAndConvert(self.filename, encode=False)
