def test_set_secondary_files(cls, secondary):
obj = cls()
i = obj.add_input(cwl.File(), id='in')
o = obj.add_output(cwl.File(), id='out')
obj.set_secondary_files('in', secondary)
obj.set_secondary_files('out', secondary)
assert i.secondary_files == secondary
assert o.secondary_files == secondary
def test_expose_except(wf):
t = CommandLineTool(id='test')
t.add_input(cwl.Int(default=10, required=True), id='x')
t.add_input(cwl.String(), id='y')
t.add_output(cwl.File(glob='something', required=True), id='out')
wf.add_step(t, expose_except=['y', 'out'])
assert wf.inputs == [WorkflowInput(id='x', label='x', type=Primitive.INT)]
assert wf.outputs == []
from sbg import cwl
with cwl.tool('tool1.cwl', 'w') as t:
t.id = 'tool1'
t.base_command = ['grep']
t.stdout = '_output_'
t.add_input(cwl.String(required=True),
'pattern',
label='pattern',
input_binding=cwl.InputBinding(shell_quote=False, position=0))
t.add_input(cwl.File(required=True),
'inFile',
label='inFile',
input_binding=cwl.InputBinding(shell_quote=False, position=1))
t.add_output(cwl.File(required=True),
'out',
label='Out',
output_binding=cwl.OutputBinding(glob='_output_'))
t.add_requirement(cwl.Docker(docker_pull='ubuntu:16.04'))
# required if we want to disable shell_quote
t.add_requirement(cwl.ShellCommand())
import pysam
from sbg import cwl
@cwl.to_tool(
inputs=dict(
bam_file=cwl.File(secondary_files=['.bai']),
bed_file=cwl.File()
),
outputs=dict(out=cwl.File(glob='gc_content.txt')),
docker='images.sbgenomics.com/filip_tubic/ubuntu1604pysam'
)
def gc_content(bam_file, bed_file):
"""Calculate GC content."""
bam_file = bam_file['path']
bed_file = bed_file['path']
bam = pysam.AlignmentFile(bam_file, 'rb')
with open('gc_content.txt', 'w') as out:
with open(bed_file) as bf:
for line in bf:
line_parts = line.strip().split()
chr = line_parts[0]
start = int(line_parts[1])
end = int(line_parts[2])
read_data = bam.fetch(chr, start, end)
total_bases = 0
gc_bases = 0
for read in read_data:
seq = read.query_sequence
total_bases += len(seq)
from sbg import cwl
import textwrap
cwl.from_bash(
label='Example tool',
inputs=dict(
HELLO="HELLO WORLD",
STR=cwl.String(),
INT=cwl.Int(),
FLOAT=cwl.Float(),
BOOL=cwl.Bool(),
ANY=cwl.Any(),
FILE=cwl.File(),
DIR=cwl.Dir(),
ENUM=cwl.Enum(['opt1', 'opt2']),
INT_OR_STR=cwl.Union([cwl.Int(), cwl.String()]),
# with default value
STR_DEF=cwl.String(default="hello"),
INT_DEF=cwl.Int(default=123),
FLOAT_DEF=cwl.Float(default=24.42),
BOOL_DEF=cwl.Bool(default=True),
ANY_DEF=cwl.Any(default="whatever"),
ENUM_DEF=cwl.Enum(['opt1', 'opt2'], default='opt2'),
INT_OR_STR_DEF=cwl.Union([cwl.Int(), cwl.String()], default=22)),
outputs=dict(out=cwl.File(glob='stdout')),
script=textwrap.dedent(r"""
echo $HELLO
echo $STR
echo $INT
echo $FLOAT
echo $BOOL
def outputs():
return dict(
out_str=cwl.String(),
out_glob_star=cwl.Array(cwl.File(), glob='*.txt'),
out_glob=cwl.File(glob="some_name")
)
def strelka(
normal_bam: cwl.File(secondary_files='.bai',
doc='Normal sample BAM or CRAM file.'),
tumor_bam: cwl.File(secondary_files='.bai',
doc='Tumor sample BAM or CRAM file.',
required=True),
reference_fasta: cwl.File(
secondary_files='.fai',
doc='samtools-indexed reference fasta file [required]'),
indel_candidates: cwl.File(
doc='Specify a VCF of candidate indel alleles. These alleles are always '
'evaluated but only reported in the output when they are inferred to '
'exist in the sample. The VCF must be tabix indexed. All indel alleles'
' must be left-shifted/normalized, any unnormalized alleles will be '
'ignored. This option may be specified more than once, multiple input '
'VCFs will be merged.',
default='None') = None,
forced_gt: cwl.File(
doc="Specify a VCF of candidate alleles. "
"These alleles are always evaluated and "
"reported even if they are unlikely to exist in the "
"sample. The VCF must be tabix indexed. All indel "
"alleles must be left-shifted/normalized, any unnormalized "
"allele will trigger a runtime error. This option may "
"be specified more than once, multiple input VCFs will "
"be merged. Note that for any SNVs provided in the VCF, "
"the SNV site will be reported (and for gVCF, excluded "
"from block compression), but the specific SNV "
"alleles are ignored.",
default='None') = None,
exome: cwl.Bool(
doc="Set options for exome or other targeted input: note in "
"particular that this flag turns off high-depth filters") = False,
call_regions: cwl.File(
doc="Optionally provide a bgzip-compressed/tabix-indexed BED "
"file containing the set of regions to call. No VCF "
"output will be provided outside of these regions. "
"The full genome will still be used to estimate statistics "
"from the input (such as expected depth per chromosome). "
"Only one BED file may be specified.",
default='Call the entire genome') = None,
scan_size_mb: cwl.Int(
doc="Maximum sequence region size (in megabases) scanned by "
"each task during genome variant calling. (default: 12)",
default=12) = 12,
region: cwl.String(
doc="Limit the analysis to one or more genome region(s) for "
"debugging purposes. If this argument is provided multiple"
" times the union of all specified regions will be analyzed. "
"All regions must be non-overlapping to get a meaningful "
"result. Examples: '--region chr20' (whole chromosome), "
"'--region chr2:100-2000 --region chr3:2500-3000' "
"(two regions)'. If this option is specified (one or more times) "
"together with the --callRegions BED file, then all "
"region arguments will be intersected with the "
"callRegions BED track.",
default='None') = None):
"""
:param normal_bam:
:param tumor_bam:
:param reference_fasta:
:param indel_candidates:
:param forced_gt:
:param exome:
:param call_regions:
:param scan_size_mb:
:param region:
:return:
"""
strelka_config_path = '/opt/bin/configureStrelkaSomaticWorkflow.py'
strelka_cmd = [strelka_config_path]
strelka_cmd += ['--normalBam', normal_bam['path']]
strelka_cmd += ['--tumorBam', tumor_bam['path']]
strelka_cmd += ['--referenceFasta', reference_fasta['path']]
strelka_cmd += ['--runDir', '.']
if indel_candidates:
strelka_cmd += ['--indelCandidates', indel_candidates['path']]
if forced_gt:
strelka_cmd += ['--forcedGT', forced_gt['path']]
if exome:
strelka_cmd += ['--exome']
if call_regions:
strelka_cmd += ['--callRegions', call_regions['path']]
strelka_cmd += ['--scanSizeMb', str(scan_size_mb)]
if region:
strelka_cmd += ['--region', region]
check_output(strelka_cmd)
check_call(['python', 'runWorkflow.py', '-m', 'local', '-j', '8'])
from sbg import cwl
from tools.doc import TOOL_DOC
from subprocess import check_call, check_output
@cwl.to_tool(outputs=dict(
stats_log=cwl.File(glob='results/stats/runStats.tsv'),
somatic_snvs=cwl.File(glob='results/variants/somatic.snvs.vcf.gz',
secondary_files='.tbi',
doc="All somatic SNVs inferred in the tumor sample"),
somatic_snvs_tbi=cwl.File(glob='results/variants/somatic.snvs.vcf.gz.tbi'),
somatic_indels=cwl.File(
glob='results/variants/somatic.indels.vcf.gz',
secondary_files='.tbi',
doc="All somatic Indels inferred in the tumor sample"),
somatic_indels_tbi=cwl.File(
glob='results/variants/somatic.indels.vcf.gz.tbi')),
docker='images.sbgenomics.com/gavrilo_andric/strelka:1',
label="Strelka 2.9.7")
def strelka(
normal_bam: cwl.File(secondary_files='.bai',
doc='Normal sample BAM or CRAM file.'),
tumor_bam: cwl.File(secondary_files='.bai',
doc='Tumor sample BAM or CRAM file.',
required=True),
reference_fasta: cwl.File(
secondary_files='.fai',
doc='samtools-indexed reference fasta file [required]'),
indel_candidates: cwl.File(
doc='Specify a VCF of candidate indel alleles. These alleles are always '
'evaluated but only reported in the output when they are inferred to '