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 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 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 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 gunzip(workflow, input_dir, **kwargs):
"""
Gunzips all gz files in directory
$ genomekey gunzip -n 'Gunzip' /path/to/dir
"""
DAG().sequence_(
add_([
INPUT(f, tags={'i': i})
for i, f in enumerate(glob.glob(os.path.join(input_dir, '*.gz')))
]), map_(unix.Gunzip), 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 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 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))
from cosmos.Workflow.models import Workflow
from cosmos.lib.ezflow.dag import DAG, add_,split_
from tools import ECHO, CAT
####################
# Workflow
####################
dag = DAG().sequence_(
add_([ ECHO(tags={'word':'hello'}), ECHO(tags={'word':'world'}) ]),
split_([('i',[1,2])],CAT)
)
dag.create_dag_img('/tmp/ex.svg')
#################
# Run Workflow
#################
WF = Workflow.start('Example 1',restart=True)
dag.add_to_workflow(WF)
WF.run()
from cosmos.Workflow.models import Workflow
from cosmos.lib.ezflow.dag import DAG, split_,add_,map_,reduce_
from tools import ECHO, CAT, WC, PASTE, Sleep
####################
# Workflow
####################
dag = DAG().sequence_(
add_([ ECHO(tags={'word':'hello'}), ECHO(tags={'word':'world'}) ]),
map_(Sleep),
split_([('i',[1,2])], CAT),
reduce_([], PASTE),
map_(WC),
)
dag.create_dag_img('/tmp/ex.svg')
#################
# Run Workflow
#################
WF = Workflow.start('Example 3',restart=True,delete_intermediates=True)
dag.add_to_workflow(WF)
WF.run()
from cosmos.lib.ezflow.dag import DAG, Split, Add, Map, Reduce
from tools import ECHO, MD5Sum
from cosmos.Workflow.cli import CLI
cli = CLI()
WF = cli.parse_args() # parses command line arguments
####################
# Workflow
####################
dag = ( DAG()
|Add| [ ECHO(tags={'word':'hello'}), ECHO(tags={'word':'world'}) ]
|Reduce| ([],MD5Sum)
)
dag.create_dag_img('/tmp/ex.svg')
#################
# Run Workflow
#################
dag.add_to_workflow(WF)
WF.run()
This workflow demonstrates branching for when you need
something more complicated than a linear step-by-step
series of stages.
cosmos.lib.ezflow.dag.DAG.branch() is the key to branching.
"""
from cosmos.Workflow.models import Workflow
from cosmos.lib.ezflow.dag import DAG
import tools
####################
# Workflow
####################
dag = (DAG().add([
tools.ECHO(tags={'word': 'hello'}),
tools.ECHO(tags={'word': 'world'})
]).split([('i', [1, 2])], tools.CAT).map(tools.WC).branch('ECHO').map(
tools.WC, 'Extra Independent Word Count'))
# Generate image
dag.create_dag_img('/tmp/ex_branch.svg')
#################
# Run Workflow
#################
WF = Workflow.start('Example Branch', restart=True)
dag.add_to_workflow(WF)
WF.run()
HOST = "smtp.server"
text = '{0} has failed at stage {1}'.format(stage.workflow, stage)
BODY = "\r\n".join(
"From: %s" % FROM,
"To: %s" % TO,
"Subject: %s" % SUBJECT,
"",
text
)
server = smtplib.SMTP(HOST)
server.sendmail(FROM, [TO], BODY)
server.quit()
####################
# Workflow
####################
from cosmos.lib.ezflow.dag import DAG, Map, Split, Add
import tools
dag = ( DAG().
add_([tools.ECHO(tags={'word': 'hello'}), tools.ECHO(tags={'word': 'world'})]).
map_(tools.FAIL) # Automatically fail
)
#################
# Run Workflow
#################
dag.add_to_workflow(WF)
WF.run()
from cosmos.Workflow.models import Workflow
from cosmos.lib.ezflow.dag import DAG, add_, split_
from tools import ECHO, CAT
####################
# Workflow
####################
dag = DAG().sequence_(
add_([ECHO(tags={'word': 'hello'}),
ECHO(tags={'word': 'world'})]), split_([('i', [1, 2])], CAT))
dag.create_dag_img('/tmp/ex.svg')
#################
# Run Workflow
#################
WF = Workflow.start('Example 1', restart=True)
dag.add_to_workflow(WF)
WF.run()