def run(self):
# Check both type of indexes exist #
if not os.path.exists(self.contigs_fasta + '.1.bt2'): self.contigs_fasta.index_bowtie()
if not os.path.exists(self.contigs_fasta + '.fai'): self.contigs_fasta.index_samtools()
# Make our options #
options = ['-p', nr_threads,
'-x', self.assembly.results.contigs_fasta,
'-1', self.sample.fwd_path,
'-2', self.sample.rev_path,
'-S', self.p.map_sam]
# We have to tell bowtie2 if they we have FASTA files instead of FASTQ #
if self.sample.format == 'fasta': options += ['-f']
# Do the mapping #
sh.bowtie2(*options)
## Create bam file, then sort it and finally index bamfile #
sh.samtools('view', '-bt', self.contigs_fasta + '.fai', self.p.map_sam, '-o', self.p.map_bam)
sh.samtools('sort', self.p.map_bam, self.p.map_s_bam.prefix_path)
sh.samtools('index', self.p.map_s_bam)
# Remove PCR duplicates #
self.remove_duplicates()
# Sort and index bam without duplicates #
sh.samtools('sort', self.p.map_smd_bam, self.p.map_smds_bam.prefix_path)
sh.samtools('index', self.p.map_smds_bam)
# Compute coverage #
sh.genomeCoverageBed('-ibam', self.p.map_smds_bam, _out=str(self.p.map_smds_coverage))
# Clean up the ones we don't need #
os.remove(self.p.map_sam)
os.remove(self.p.map_bam)
os.remove(self.p.map_smd_bam)
def map(self,threads=nr_threads):
"""Maps reads from self.pool to self.assembly using bowtie2. PCR
Duplicates are afterwards removed using MarkDuplicates. BEDTools is
used to determine coverage."""
# Check indexes #
if not os.path.exists(self.contigs + '.1.bt2'):
raise(Exception('Bowtie2 index file not created, run index() first'))
if not os.path.exists(self.contigs + '.fai'):
raise(Exception('Samtools index file not created, run index() first'))
# Do the mapping #
sh.bowtie2('-p', threads, '-x', self.contigs, '-1', self.pool.fwd, '-2', self.pool.rev, '-S', self.p.map_sam)
# Create bam, sort and index bamfile #
sh.samtools('view', '-bt', self.contigs + '.fai', self.p.map_sam, '-o', self.p.map_bam)
sh.samtools('sort', self.p.map_bam, self.p.map_s_bam.prefix_path)
sh.samtools('index', self.p.map_s_bam)
# Remove PCR duplicates #
self.remove_duplicates()
# Sort and index bam without duplicates #
sh.samtools('sort', self.p.map_smd_bam, self.p.map_smds_bam.prefix_path)
sh.samtools('index', self.p.map_smds_bam)
# Compute coverage #
sh.genomeCoverageBed('-ibam', self.p.map_smds_bam, _out=self.p.map_smds_coverage)
# Clean up #
os.remove(self.p.map_sam)
os.remove(self.p.map_bam)
os.remove(self.p.map_smd_bam)
def index_fasta(fasta, compress):
"""Index and optionally compress a fasta file.
\b
Examples:
bionorm index_fasta Medicago_truncatula/jemalong_A17.gnm5.ann1.FAKE/medtr.jemalong_A17.gnm5.FAKE.genome_main.fna
"""
from sh import bgzip # isort:skip
from sh import samtools # isort:skip
target = Path(fasta)
if len(target.suffixes) < 1:
error_message = f"Target {target} does not have a file extension."
logger.error(error_message)
sys.exit(1)
if target.suffix.lstrip(".") in COMPRESSED_TYPES:
logger.error(f"Uncompress {target} befor indexing.")
sys.exit(1)
if target.suffix.lstrip(".") not in FASTA_TYPES:
logger.error(
f"File {target} does not have a recognized FASTA extension.")
sys.exit(1)
if compress:
output = bgzip(["-f", "--index", str(target)])
target = Path(target.parent) / f"{target.name}.gz"
output = samtools(["faidx", str(target)])
return target
def main():
'''
Call `samtools view` on the input file and split into fastqs by RNAME column.
'''
raw_args = docopt(__doc__)
scheme = Schema({
'<samfile>' : str,
'--outdir' : str})
parsed_args = scheme.validate(raw_args)
outdir = parsed_args['--outdir']
if not os.path.exists(outdir):
os.mkdir(outdir)
infile = parsed_args['<samfile>']
view = str(sh.samtools('view', infile, S=True)) if infile.endswith('.sam') else str(sh.samtools('view', infile))
get_seqs_by_ctg(outdir, view)
return 0
def _merge_condition(in_files, condition):
"""
merge all of the bam files from a condition together
as recomended in the MACS manual
"""
condition_files = [filename for filename in in_files if
condition in filename]
if not condition_files:
return None
condition_filename = os.path.join(os.path.dirname(condition_files[1]),
condition + "_merged.bam")
sorted_prefix = remove_suffix(condition_filename) + ".sorted"
sorted_filename = sorted_prefix + ".bam"
if file_exists(sorted_filename):
return sorted_filename
sh.samtools("merge", condition_filename, condition_files)
sh.samtools("sort", condition_filename, sorted_prefix)
sh.samtools("index", sorted_filename)
return sorted_filename
def _merge_condition(in_files, condition):
"""
merge all of the bam files from a condition together
as recomended in the MACS manual
"""
condition_files = [
filename for filename in in_files if condition in filename
]
if not condition_files:
return None
condition_filename = os.path.join(os.path.dirname(condition_files[1]),
condition + "_merged.bam")
sorted_prefix = remove_suffix(condition_filename) + ".sorted"
sorted_filename = sorted_prefix + ".bam"
if file_exists(sorted_filename):
return sorted_filename
sh.samtools("merge", condition_filename, condition_files)
sh.samtools("sort", condition_filename, sorted_prefix)
sh.samtools("index", sorted_filename)
return sorted_filename
def index_samtools(self):
"""Create an index on the fasta file compatible with samtools"""
sh.samtools('faidx', self.path)
return FilePath(self.path + '.fai')
def index(self):
"""Create two indexes. For both bowtie2 and samtools on the contigs fasta file."""
sh.bowtie2_build(self.contigs_fasta, self.contigs_fasta)
sh.samtools('faidx', self.contigs_fasta)
def zero_coverage_positions(bam_file, ref_file):
pileup = sh.samtools('mpileup', bam_file, f=ref_file, _iter=True)
return map(compose(int, second, unicode.split), pileup)
def index_samtools(self):
"""Create an index on the fasta file compatible with samtools."""
sh.samtools('faidx', self.path)
return FilePath(self.path + '.fai')
