def json_(workflow,input_dict,**kwargs):
"""
Input file is a json of the following format:
[
{
'chunk': 001,
'library': 'LIB-1216301779A',
'sample_name': '1216301779A',
'platform': 'ILLUMINA',
'platform_unit': 'C0MR3ACXX.001'
'pair': 0, #0 or 1
'path': '/path/to/fastq'
},
{..}
]
"""
input_json = json.load(open(input_dict,'r'))
inputs = [ INPUT(name='fastq.gz',path=i['path'],fmt='fastq.gz',tags=i,stage_name='Load Input Fastqs') for i in input_json ]
DAG(ignore_stage_name_collisions=True).sequence_(
add_(inputs),
Pipeline(),
configure(wga_settings),
add_run(workflow)
)
def bam(workflow,input_bam,input_bam_list,**kwargs):
"""
Input file is a bam with properly annotated readgroups.
*** Note that this workflow assumes the bam header is ***
*** also properly annotated with the correct readgroups! ***
Example usage:
$ genomekey bam -n 'Bam to VCF Workflow 1' input_bam.bam
$ echo "dir/sample1.bam" > /tmp/bam.list
$ echo "dir/sample2.bam" >> /tmp/bam.list
$ genomekey bam -n 'Bam to VCF 2' -li /tmp/bam.list
"""
# capture and pedigree_file are used in main()
input_bams = input_bam_list.read().strip().split('\n') if input_bam_list else []
if input_bam:
input_bams.append(input_bam.name)
dag = DAG(ignore_stage_name_collisions=True)
Bam2Fastq(workflow,dag,wga_settings,input_bams)
dag.sequence_(
Pipeline(),
configure(wga_settings),
add_run(workflow)
)
def json_(workflow, input_dict, **kwargs):
"""
Input file is a json of the following format:
[
{
'chunk': 001,
'library': 'LIB-1216301779A',
'sample_name': '1216301779A',
'platform': 'ILLUMINA',
'platform_unit': 'C0MR3ACXX.001'
'pair': 0, #0 or 1
'path': '/path/to/fastq'
},
{..}
]
"""
input_json = json.load(open(input_dict, 'r'))
inputs = [
INPUT(name='fastq.gz',
path=i['path'],
fmt='fastq.gz',
tags=i,
stage_name='Load Input Fastqs') for i in input_json
]
DAG(ignore_stage_name_collisions=True).sequence_(add_(inputs), Pipeline(),
configure(wga_settings),
add_run(workflow))
def bam(workflow, input_bam, input_bam_list, **kwargs):
"""
Input file is a bam with properly annotated readgroups.
*** Note that this workflow assumes the bam header is ***
*** also properly annotated with the correct readgroups! ***
Example usage:
$ genomekey bam -n 'Bam to VCF Workflow 1' input_bam.bam
$ echo "dir/sample1.bam" > /tmp/bam.list
$ echo "dir/sample2.bam" >> /tmp/bam.list
$ genomekey bam -n 'Bam to VCF 2' -li /tmp/bam.list
"""
# capture and pedigree_file are used in main()
input_bams = input_bam_list.read().strip().split(
'\n') if input_bam_list else []
if input_bam:
input_bams.append(input_bam.name)
dag = DAG(ignore_stage_name_collisions=True)
Bam2Fastq(workflow, dag, wga_settings, input_bams)
dag.sequence_(Pipeline(), configure(wga_settings), add_run(workflow))
def json_somatic(workflow,input_dict,**kwargs):
"""
Input file is a json of the following format:
[
{
"chunk": "001",
"library": "LIB-1216301779A",
"platform": "ILLUMINA",
"platform_unit": "C0MR3ACXX.001",
"rgid": "BC18-06-2013",
"sample_name": "BC18-06-2013LyT_S5_L001",
"pair": "1",
"path": "/path/to/fastq.gz",
"sample_type": "normal or tumor"
},
{..}
]
"""
input_json = json.load(open(input_dict,'r'))
inputs = [ INPUT(name='fastq.gz',path=i['path'],fmt='fastq.gz',tags=i,stage_name='Load Input Fastqs') for i in input_json ]
DAG(ignore_stage_name_collisions=True).sequence_(
add_(inputs),
Pipeline_Somatic(),
configure(wga_settings),
add_run(workflow)
)
def json_somatic(workflow, input_dict, **kwargs):
"""
Input file is a json of the following format:
[
{
"chunk": "001",
"library": "LIB-1216301779A",
"platform": "ILLUMINA",
"platform_unit": "C0MR3ACXX.001",
"rgid": "BC18-06-2013",
"sample_name": "BC18-06-2013LyT_S5_L001",
"pair": "1",
"path": "/path/to/fastq.gz",
"sample_type": "normal or tumor"
},
{..}
]
"""
input_json = json.load(open(input_dict, 'r'))
inputs = [
INPUT(name='fastq.gz',
path=i['path'],
fmt='fastq.gz',
tags=i,
stage_name='Load Input Fastqs') for i in input_json
]
DAG(ignore_stage_name_collisions=True).sequence_(add_(inputs),
Pipeline_Somatic(),
configure(wga_settings),
add_run(workflow))
def json_local(workflow,input_dict,**kwargs):
"""
Input is a folder where each file is a json of the following format:
[
{
'library': 'LIB-1216301779A',
'sample_name': '1216301779A',
'platform': 'ILLUMINA',
'platform_unit': 'C0MR3ACXX.001'
'pair':1
'path': '/path/to/fastq'
},
{
'library': 'LIB-1216301779A',
'sample_name': '1216301779A',
'platform': 'ILLUMINA',
'platform_unit': 'C0MR3ACXX.001'
'pair':2
'path': '/path/to/fastq'..}
]
"""
dirList=os.listdir(input_dict)
for files in dirList:
print input_dict+files
input_json = json.load(open(input_dict+files,'r'))
inputs = [ INPUT(name='fastq.gz',path=i['path'],fmt='fastq.gz',tags=i,stage_name='Load Input Fastqs') for i in input_json ]
for i in inputs:
print i
DAG(ignore_stage_name_collisions=True).sequence_(
add_(inputs),
Pipeline_local(),
configure(wga_settings),
add_run(workflow)
)
def downdbs(workflow,**kwargs):
"""
Download all annotation databases
"""
DAG().sequence_(
add_([ annovarext.DownDB(tags={'build':'hg19','dbname':db}) for db in annovarext.get_db_names() ]),
configure(wga_settings),
add_run(workflow)
)
def downdbs(workflow, **kwargs):
"""
Download all annotation databases
"""
DAG().sequence_(
add_([
annovarext.DownDB(tags={
'build': 'hg19',
'dbname': db
}) for db in annovarext.get_db_names()
]), configure(wga_settings), add_run(workflow))
def fastq_(workflow,input_dict,output_dict,output_json,**kwargs):
json_fastq_to_split=json_creator.json_out(input_dict,output_dict)
input_json = json.load(open(json_fastq_to_split,'r'))
inputs = [ INPUT(name='fastq.gz',path=i['gz_path'],fmt='fastq.gz',tags=i,stage_name='Load Input Fastqs') for i in input_json ]
DAG(ignore_stage_name_collisions=True).sequence_(
add_(inputs),
Pipeline_split(),
configure(wga_settings),
add_run(workflow)
)
def upload_(workflow,bucket,project,out_dict,**kwargs):
project_folder=join(out_dict,project.replace(" ", "_"))
if not os.path.exists(project_folder):
os.makedirs(project_folder)
json_fastq_to_upload=s3_Bucket.getList(bucket,project,out_dict)
input_json = json.load(open(json_fastq_to_upload,'r'))
inputs = [ INPUT(name='fastq.gz',path=i['gz_path'],fmt='fastq.gz',tags=i,stage_name='Load Input Fastqs') for i in input_json ]
DAG(ignore_stage_name_collisions=True).sequence_(
add_(inputs),
Pipeline_upload(),
configure(wga_settings),
add_run(workflow)
)
def fastq_(workflow, input_dict, output_dict, output_json, **kwargs):
json_fastq_to_split = json_creator.json_out(input_dict, output_dict)
input_json = json.load(open(json_fastq_to_split, 'r'))
inputs = [
INPUT(name='fastq.gz',
path=i['gz_path'],
fmt='fastq.gz',
tags=i,
stage_name='Load Input Fastqs') for i in input_json
]
DAG(ignore_stage_name_collisions=True).sequence_(add_(inputs),
Pipeline_split(),
configure(wga_settings),
add_run(workflow))
def anno(workflow, input_file, input_file_list, file_format='vcf', **kwargs):
"""
Annotates all files in input_Files
$ genomekey anno -n 'My Annotation Workflow #1' file1.vcf file2.vcf
"""
input_files = input_file_list.read().strip().split(
'\n') if input_file_list else []
if input_file:
input_files.append(input_file.name)
print('annotating {0}'.format(', '.join(input_files)), file=sys.stderr)
DAG().sequence_(
add_([
INPUT(input_file, tags={'vcf': i})
for i, input_file in enumerate(input_files)
]), massive_annotation, configure(wga_settings), add_run(workflow))
def anno(workflow,input_file,input_file_list,file_format='vcf',**kwargs):
"""
Annotates all files in input_Files
$ genomekey anno -n 'My Annotation Workflow #1' file1.vcf file2.vcf
"""
input_files = input_file_list.read().strip().split('\n') if input_file_list else []
if input_file:
input_files.append(input_file.name)
print >> sys.stderr, 'annotating {0}'.format(', '.join(input_files))
DAG().sequence_(
add_([ INPUT(input_file,tags={'vcf':i}) for i,input_file in enumerate(input_files) ]),
massive_annotation,
configure(wga_settings),
add_run(workflow)
)
def upload_(workflow, bucket, project, out_dict, **kwargs):
project_folder = join(out_dict, project.replace(" ", "_"))
if not os.path.exists(project_folder):
os.makedirs(project_folder)
json_fastq_to_upload = s3_Bucket.getList(bucket, project, out_dict)
input_json = json.load(open(json_fastq_to_upload, 'r'))
inputs = [
INPUT(name='fastq.gz',
path=i['gz_path'],
fmt='fastq.gz',
tags=i,
stage_name='Load Input Fastqs') for i in input_json
]
DAG(ignore_stage_name_collisions=True).sequence_(add_(inputs),
Pipeline_upload(),
configure(wga_settings),
add_run(workflow))
def Bam2Fastq(workflow, dag, settings, input_bams):
if len(input_bams) == 0:
raise WorkflowException, 'At least 1 BAM input required'
dag.sequence_(
sequence_(
*[
sequence_(
add_([ INPUT(input_bam, tags={'input':os.path.basename(input_bam)})],stage_name="Load Input Bams"),
split_([('rgid',_inputbam2rgids(input_bam))],pipes.FilterBamByRG_To_FastQ)
)
for input_bam in input_bams
],
combine=True
),
split_([('pair',[1,2])],genomekey_scripts.SplitFastq),
configure(settings),
add_run(workflow,finish=False),
).add_(list(_splitfastq2inputs(dag)))
return dag
def Bam2Fastq(workflow, dag, settings, input_bams):
if len(input_bams) == 0:
raise WorkflowException, 'At least 1 BAM input required'
dag.sequence_(
sequence_(*[
sequence_(
add_([
INPUT(input_bam,
tags={'input': os.path.basename(input_bam)})
],
stage_name="Load Input Bams"),
split_([('rgid', _inputbam2rgids(input_bam))],
pipes.FilterBamByRG_To_FastQ))
for input_bam in input_bams
],
combine=True),
split_([('pair', [1, 2])], genomekey_scripts.SplitFastq),
configure(settings),
add_run(workflow, finish=False),
).add_(list(_splitfastq2inputs(dag)))
return dag
def json_local(workflow, input_dict, **kwargs):
"""
Input is a folder where each file is a json of the following format:
[
{
'library': 'LIB-1216301779A',
'sample_name': '1216301779A',
'platform': 'ILLUMINA',
'platform_unit': 'C0MR3ACXX.001'
'pair':1
'path': '/path/to/fastq'
},
{
'library': 'LIB-1216301779A',
'sample_name': '1216301779A',
'platform': 'ILLUMINA',
'platform_unit': 'C0MR3ACXX.001'
'pair':2
'path': '/path/to/fastq'..}
]
"""
dirList = os.listdir(input_dict)
for files in dirList:
print(input_dict + files)
input_json = json.load(open(input_dict + files, 'r'))
inputs = [
INPUT(name='fastq.gz',
path=i['path'],
fmt='fastq.gz',
tags=i,
stage_name='Load Input Fastqs') for i in input_json
]
for i in inputs:
print(i)
DAG(ignore_stage_name_collisions=True).sequence_(
add_(inputs), Pipeline_local(), configure(wga_settings),
add_run(workflow))