def open_possibly_compressed_file(filename):
""" Read a file. It might be compressed. """
#if isinstance(filename, file):
# return filename #Whatever
#peek, f = peek_and_pipe(open_possibly_remote_file(filename), 4)
from nesoni import sam
if sam.is_bam(filename):
return sam.open_bam(filename)
f = open(filename, 'rb')
peek = f.read(4)
f.close()
#f = open(filename,'rb')
if peek.startswith('\x1f\x8b'):
#command = 'gunzip'
return gzip.open(filename, 'rb')
elif peek.startswith('BZh'):
#command = 'bunzip2'
return bz2.BZFile(filename, 'rb')
else:
#command = None
return open(filename, 'rb')
def open_possibly_compressed_file(filename):
""" Read a file. It might be compressed. """
#if isinstance(filename, file):
# return filename #Whatever
#peek, f = peek_and_pipe(open_possibly_remote_file(filename), 4)
from nesoni import sam
if sam.is_bam(filename):
return sam.open_bam(filename)
f = open(filename,'rb')
peek = f.read(4)
f.close()
#f = open(filename,'rb')
if peek.startswith('\x1f\x8b'):
#command = 'gunzip'
return gzip.open(filename, 'rb')
elif peek.startswith('BZh'):
#command = 'bunzip2'
return bz2.BZFile(filename, 'rb')
else:
#command = None
return open(filename, 'rb')
def get_compression_type(filename):
if hasattr(filename, 'read'):
return 'none' #It's already file-like
from nesoni import sam
f = open(filename,'rb')
peek = f.read(4)
f.close()
if peek.startswith('\x1f\x8b'): #gzip format
if sam.is_bam(filename): #it might be a BAM
return 'bam'
return 'gzip'
elif peek.startswith('BZh'): #bzip2 format
return 'bzip2'
else:
return 'none'
def get_compression_type(filename):
if hasattr(filename, 'read'):
return 'none' #It's already file-like
from nesoni import sam
f = open(filename, 'rb')
peek = f.read(4)
f.close()
if peek.startswith('\x1f\x8b'): #gzip format
if sam.is_bam(filename): #it might be a BAM
return 'bam'
return 'gzip'
elif peek.startswith('BZh'): #bzip2 format
return 'bzip2'
else:
return 'none'
def run(self):
workspace = working_directory.Working(self.output_dir)
workspace.setup_reference(self.reference)
workspace.update_param(snp_cost=self.snp_cost)
# assert os.path.exists(self.reference), 'Reference file does not exist'
# reference_filename = workspace._object_filename('reference.fa')
# if os.path.exists(reference_filename):
# os.unlink(reference_filename)
# os.symlink(os.path.relpath(self.reference, self.output_dir), reference_filename)
bam_filename = io.abspath(self.output_dir, "alignments.bam")
bam_prefix = io.abspath(self.output_dir, "alignments")
if sam.is_bam(self.input):
sort_input_filename = self.input
temp_filename = None
else:
temp_filename = io.abspath(self.output_dir, "temp.bam")
sort_input_filename = temp_filename
writer = io.Pipe_writer(temp_filename, ["samtools", "view", "-S", "-b", "-"])
f = open(self.input, "rb")
while True:
data = f.read(1 << 20)
if not data:
break
writer.write(data)
writer.close()
f.close()
grace.status("Sort")
# io.execute([
# 'samtools', 'sort', '-n', sort_input_filename, bam_prefix
# ])
sam.sort_bam(sort_input_filename, bam_prefix, by_name=True)
if temp_filename is not None:
os.unlink(temp_filename)
grace.status("")
def run(self):
workspace = working_directory.Working(self.output_dir)
workspace.setup_reference(self.reference)
workspace.update_param(snp_cost = self.snp_cost)
#assert os.path.exists(self.reference), 'Reference file does not exist'
#reference_filename = workspace._object_filename('reference.fa')
#if os.path.exists(reference_filename):
# os.unlink(reference_filename)
#os.symlink(os.path.relpath(self.reference, self.output_dir), reference_filename)
bam_filename = io.abspath(self.output_dir, 'alignments.bam')
bam_prefix = io.abspath(self.output_dir, 'alignments')
if sam.is_bam(self.input):
sort_input_filename = self.input
temp_filename = None
else:
temp_filename = io.abspath(self.output_dir, 'temp.bam')
sort_input_filename = temp_filename
writer = io.Pipe_writer(temp_filename, ['samtools', 'view', '-S', '-b', '-'])
f = open(self.input, 'rb')
while True:
data = f.read(1<<20)
if not data: break
writer.write(data)
writer.close()
f.close()
grace.status('Sort')
#io.execute([
# 'samtools', 'sort', '-n', sort_input_filename, bam_prefix
#])
sam.sort_bam(sort_input_filename, bam_prefix, by_name=True)
if temp_filename is not None:
os.unlink(temp_filename)
grace.status('')
def run(self):
workspace = working_directory.Working(self.output_dir)
workspace.setup_reference(self.reference)
workspace.update_param(snp_cost = self.snp_cost)
#assert os.path.exists(self.reference), 'Reference file does not exist'
#reference_filename = workspace._object_filename('reference.fa')
#if os.path.exists(reference_filename):
# os.unlink(reference_filename)
#os.symlink(os.path.relpath(self.reference, self.output_dir), reference_filename)
bam_filename = io.abspath(self.output_dir, 'alignments.bam')
bam_prefix = io.abspath(self.output_dir, 'alignments')
if sam.is_bam(self.input):
sort_input_filename = self.input
temp_filename = None
else:
temp_filename = io.abspath(self.output_dir, 'temp.bam')
sort_input_filename = temp_filename
writer = io.Pipe_writer(temp_filename, ['samtools', 'view', '-S', '-b', '-'])
f = open(self.input, 'rb')
while True:
data = f.read(1<<20)
if not data: break
writer.write(data)
writer.close()
f.close()
grace.status('Sort')
io.execute([
'samtools', 'sort', '-n', sort_input_filename, bam_prefix
])
if temp_filename is not None:
os.unlink(temp_filename)
grace.status('')
def run(self):
bams = [ ]
reference = None
reference2 = None
extra = [ ]
for sample in self.samples:
if sam.is_bam(sample):
bams.append(sample)
elif os.path.isdir(sample):
working = working_directory.Working(sample,True)
bams.append( working.get_filtered_sorted_bam() )
extra.append( '##sampleTags=' + ','.join(working.get_tags()) )
if reference2 is None:
reference2 = working.get_reference().reference_fasta_filename()
elif io.is_sequence_file(sample):
assert reference is None, 'Only one reference FASTA file allowed.'
reference = sample
if reference is None:
reference = reference2
if reference is None:
raise grace.Error('No reference FASTA file given.')
with nesoni.Stage() as stage:
tempspace = stage.enter( workspace.tempspace() )
if self.depth_limit:
with nesoni.Stage() as stage2:
for i in xrange(len(bams)):
sam.Bam_depth_limit(
tempspace/('%d'%i),
bams[i],
depth=self.depth_limit
).process_make(stage2)
bams[i] = tempspace/('%d.bam'%i)
# FreeBayes claims to handle multiple bams, but it doesn't actually work
if len(bams) > 1:
sam.Bam_merge(tempspace/'merged', bams=bams, index=False).run()
bams = [ tempspace/'merged.bam' ]
command = [
'freebayes',
'-f', reference,
'--ploidy',str(self.ploidy),
'--pvar',str(self.pvar),
] + self.freebayes_options + bams
self.log.log('Running: '+' '.join(command)+'\n')
f_out = stage.enter( open(self.prefix+'.vcf','wb') )
f_in = stage.enter( io.pipe_from(command) )
done_extra = False
for line in f_in:
if not done_extra and not line.startswith('##'):
for extra_line in extra:
f_out.write(extra_line+'\n')
done_extra = True
f_out.write(line)
index_vcf(self.prefix+'.vcf')
def run(self):
bams = []
reference = None
reference2 = None
extra = []
for sample in self.samples:
if sam.is_bam(sample):
bams.append(sample)
elif os.path.isdir(sample):
working = working_directory.Working(sample, True)
bams.append(working.get_filtered_sorted_bam())
extra.append('##sampleTags=' + ','.join(working.get_tags()))
if reference2 is None:
reference2 = working.get_reference(
).reference_fasta_filename()
elif io.is_sequence_file(sample):
assert reference is None, 'Only one reference FASTA file allowed.'
reference = sample
if reference is None:
reference = reference2
if reference is None:
raise grace.Error('No reference FASTA file given.')
with nesoni.Stage() as stage:
tempspace = stage.enter(workspace.tempspace())
if self.depth_limit:
with nesoni.Stage() as stage2:
for i in xrange(len(bams)):
sam.Bam_depth_limit(
tempspace / ('%d' % i),
bams[i],
depth=self.depth_limit).process_make(stage2)
bams[i] = tempspace / ('%d.bam' % i)
# FreeBayes claims to handle multiple bams, but it doesn't actually work
if len(bams) > 1:
sam.Bam_merge(tempspace / 'merged', bams=bams,
index=False).run()
bams = [tempspace / 'merged.bam']
command = [
'freebayes',
'-f',
reference,
'--ploidy',
str(self.ploidy),
'--pvar',
str(self.pvar),
] + self.freebayes_options + bams
self.log.log('Running: ' + ' '.join(command) + '\n')
f_out = stage.enter(open(self.prefix + '.vcf', 'wb'))
f_in = stage.enter(io.pipe_from(command))
done_extra = False
for line in f_in:
if not done_extra and not line.startswith('##'):
for extra_line in extra:
f_out.write(extra_line + '\n')
done_extra = True
f_out.write(line)
index_vcf(self.prefix + '.vcf')