def merge_vcf(b: hb.batch.Batch,
gvcf_list: List = None,
output_vcf_name: str = None,
merge_vcfs_img: str = None,
memory: int = 3,
out_dir: str = None,
storage: int = None):
# Combine multiple VCFs or GVCFs from scattered HaplotypeCaller runs
docker_image = merge_vcfs_img if merge_vcfs_img else 'us.gcr.io/broad-gotc-prod/genomes-in-the-cloud:2.4.3-1564508330'
outname = output_vcf_name + '.g.vcf.gz'
# disk_size = bytes_to_gb((inputs_vcfs_list * 2.5)) + 10
merge_vcf_i = ''
for line in gvcf_list:
input_gvcf = b.read_input(line)
merge_vcf_i += f'I={input_gvcf} \t'
merge_vcfs = b.new_job(name=output_vcf_name)
merge_vcfs.image(docker_image)
merge_vcfs.memory(f'{memory}Gi')
merge_vcfs.storage(f'{storage}Gi')
merge_vcfs.command(f'java -Xms2000m -jar /usr/gitc/picard.jar \
MergeVcfs \
{merge_vcf_i} \
O={outname}')
merge_vcfs.command(f'mv {outname} {merge_vcfs.ofile}')
b.write_output(merge_vcfs.ofile,
f'{out_dir}/merged-gvcf/{output_vcf_name}/{outname}')
return merge_vcfs
def merge_vcf(b: hb.batch.Batch,
gvcf_list: List = None,
output_vcf_name: str = None,
merge_vcfs_img: str = None,
memory: int = 3,
out_dir: str = None,
storage: int = None):
"""
Combine multiple VCFs or GVCFs from scattered HaplotypeCaller runs
:param b: batch
:param gvcf_list: list of GVCF files to merge
:param output_vcf_name: output GVCF name
:param merge_vcfs_img: image to use for the job
:param storage: Storage to use fo the job
:param out_dir: output directory
:param memory: job memory
:return:
"""
docker_image = merge_vcfs_img if merge_vcfs_img else\
'us.gcr.io/broad-gotc-prod/genomes-in-the-cloud:2.4.3-1564508330'
outname = output_vcf_name + '.g.vcf.gz'
merge_vcf_i = ''
for line in gvcf_list:
input_gvcf = b.read_input(line)
merge_vcf_i += f'I={input_gvcf} \t'
merge_vcfs = b.new_job(name=output_vcf_name)
merge_vcfs.image(docker_image)
merge_vcfs.memory(f'{memory}Gi')
merge_vcfs.storage(f'{storage}Gi')
merge_vcfs.command(f'java -Xms2000m -jar /usr/gitc/picard.jar \
MergeVcfs \
{merge_vcf_i} \
O={outname}')
merge_vcfs.command(f'mv {outname} {merge_vcfs.ofile}')
b.write_output(merge_vcfs.ofile,
f'{out_dir}/merged-gvcf/{output_vcf_name}/{outname}')
return merge_vcfs
def cram_to_bam(b: hb.batch.Batch,
input_cram_file: str = None,
ref_fasta: str = None,
ref_dict: str = None,
ref_ind: str = None,
bam_out_name: str = None,
memory: int = 15,
samtools_image: str = None,
out_dir: str = None):
docker_image = samtools_image if samtools_image else 'gcr.io/genomics-tools/samtools'
out_bam_name = bam_out_name + '.bam'
output_bam_size: float = bytes_to_gb(input_cram_file) / 0.40
ref_size: float = bytes_to_gb(ref_fasta) + bytes_to_gb(ref_ind)
disk_size: int = round(
bytes_to_gb(input_cram_file) + output_bam_size + ref_size) + 25
job_memory = str(memory) + 'Gi'
job_storage = str(disk_size) + 'Gi'
crams_to_bams = b.new_job(name=out_bam_name)
in_cram = b.read_input(input_cram_file)
fasta = b.read_input_group(**{
'fasta': ref_fasta,
'fasta.fai': ref_ind,
'dict': ref_dict
})
crams_to_bams.memory(job_memory)
crams_to_bams.image(docker_image)
crams_to_bams.storage(job_storage)
crams_to_bams.command(
f'samtools view -b -T {fasta.fasta} -o {out_bam_name} {in_cram}')
crams_to_bams.command(f'samtools index {out_bam_name}')
crams_to_bams.command(f'mv {out_bam_name} {crams_to_bams.bamout}')
crams_to_bams.command(f'mv {out_bam_name}.bai {crams_to_bams.bamind}')
b.write_output(crams_to_bams.bamout, f'{out_dir}/BAMS/{out_bam_name}')
b.write_output(crams_to_bams.bamind, f'{out_dir}/BAMS/{out_bam_name}.bai')
return crams_to_bams