def merge_vcf_wrap(args):
"""
Merge VCFS
"""
vcfs = args.vcfs
in_file = input_file.InputFile(args.input_file, args.output_directory)
#merge_cnvs.merge_cnvs_clusters(vcfs, args.temp_dir,args.output_directory)
merge_cnvs.merge_cnvs_indiv(vcfs, args.temp_dir, args.output_directory,
in_file)
def pilon_align_wrap(args):
in_file = input_file.InputFile(args.input_file, args.input_directory)
try:
os.mkdir(args.temp_dir)
except OSError:
pass
try:
os.mkdir(args.output_directory)
except OSError:
pass
# Align reads only if they don't already exist.
# Then,run pilon to fix the reads removing contigs that are bad.
# If PE and SE files are present run pilon to fix the contigs otherwise just filter.
for sample in in_file:
temp_fasta = scaffold.scaffold_filter(sample.scaffolds, args.temp_dir,
sample.min_contig_length)
align.align_reads(sample, args.temp_dir, temp_fasta, skip=True)
pilon.run_pilon_contigs(sample, args.temp_dir, temp_fasta)
# transfer the Pilon aligned bam files somewhere
bwa_output = align.simple_bwa(sample, args.temp_dir,
args.reference_file)
shutil.copy(
bwa_output,
os.path.join(args.output_directory, os.path.basename(bwa_output)))
shutil.copy(
bwa_output + ".bai",
os.path.join(args.output_directory,
os.path.basename(bwa_output) + ".bai"))
shutil.copy(
sample.pilon_fasta,
os.path.join(args.output_directory,
os.path.basename(sample.pilon_fasta)))
for pairs in sample.paired_end_list:
shutil.copy(
pairs[0],
os.path.join(args.output_directory,
os.path.basename(pairs[0])))
shutil.copy(
pairs[1],
os.path.join(args.output_directory,
os.path.basename(pairs[1])))
samtools_faidx = " samtools faidx {0}".format(
os.path.join(args.output_directory,
os.path.basename(sample.pilon_fasta)))
subprocess.check_call(samtools_faidx, shell=True)
# Move the pilon bams so they are stored for access
basename = sample.bam_file.split(".bam")[0]
shutil.copy(sample.bam_file, basename + ".pilon.bam")
shutil.copy(sample.bam_file + '.bai', basename + ".pilon.bam.bai")
def genotype_inversions(input_file, input_directory, output_directory, inversions):
"""
Genotyping inversions using short-read data..
"""
in_file = input_file.InputFile(args.input_file, args.input_directory)
for sample in in_file:
inversion_file = os.path.join(output_directory, sample.samples_name + ".inv")
print(inversion_file)
align.align_reads(sample,args.temp_dir, reference_file, align_to_ref=True,skip=True)
print(sample.bam_file)
# Read in file.
# Call each CNV using short-read data.
sys.exit(1)
def cnv_call_wrap(args):
"""
CNV call
"""
in_file = input_file.InputFile(args.input_file, args.input_directory)
try:
os.mkdir(args.temp_dir)
except OSError:
pass
try:
os.mkdir(args.output_directory)
except OSError:
pass
for sample in in_file:
bwa_input = os.path.join(
args.input_directory,
sample.samples_name + ".align_to_ref.sorted.bam")
fasta_reference = os.path.join(args.input_directory,
sample.samples_name + ".fasta")
cnv.call_cnvs(sample, args.mapping_quality, bwa_input,
args.output_directory, fasta_reference)
def genotype_cnvs_wrap(args):
"""
Genotype CNVs
"""
in_file = input_file.InputFile(args.input_file, args.input_directory)
try:
os.mkdir(args.temp_dir)
except OSError:
pass
try:
os.mkdir(args.output_directory)
except OSError:
pass
reference_file = args.reference_file
for sample in in_file:
# Have to align
break_point_folder = os.path.join(args.output_directory,
sample.samples_name, "breakpoints")
align.align_reads(sample, args.temp_dir, reference_file, skip=True)
input_cnvs = (os.path.join(args.input_directory,
sample.samples_name + ".cnv"))
# Index and create fastas from fastq.
fasta.index_fasta(sample, args.temp_dir)
cnvs = cnv_calls.CNVs(input_cnvs, sample.bam_file, sample.fasta_one,
sample.fasta_two, break_point_folder)
for i in range(len(cnvs)):
# Skip chromosome M for now.
if cnvs.input_rows[i]._chrom1 != "chrM" and cnvs.input_rows[
i]._variant_type != "misaligned_sv":
cnvs.extract_windowed_bam_reads(i)
# Write VCF file
vcf_output = open(
os.path.join(args.output_directory, sample.samples_name + ".vcf"),
"w")
vcf.write_vcf_header(sample.samples_name, vcf_output)
for i in range(len(cnvs)):
if cnvs.input_rows[i]._chrom1 != "chrM" and cnvs.input_rows[
i]._variant_type != "misaligned_sv":
vcf.write_vcf_row(cnvs.input_rows[i], vcf_output)