Name: discohorts
Owner: Hammer Lab
Description: Generate Cohorts based on Epidisco and/or Biokepi results
Created: 2017-02-24 21:49:38.0
Updated: 2017-03-24 17:04:10.0
Pushed: 2017-09-25 18:27:17.0
Homepage: null
Size: 46
Language: Python
GitHub Committers
User | Most Recent Commit | # Commits |
---|
Other Committers
User | Most Recent Commit | # Commits |
---|
Generate Cohorts based on Epidisco and/or Biokepi results
utils.data import load_cohort
discohorts import Discohort
rt = load_cohort(min_tumor_mtc=0, min_normal_mtc=0)
_dirs = ["/nfs-pool-{}/biokepi/".format(i) for i in range(2, 17)]
_results_dir = "/nfs-pool/biokepi/results"
rt = Discohort(
cohort,
biokepi_work_dirs=work_dirs,
dest_results_dir=dest_results_dir)
rt.add_epidisco_pipeline(
pipeline_name="epidisco_1")
stomize the normal BAM input.
rt.add_epidisco_pipeline(
pipeline_name="epidisco_2",
config=EpidiscoConfig(cohort,
arg_normal_input=lambda patient: patient.normal_sample.bam_path_dna))
stomize the normal BAM input and the run name.
rt.add_epidisco_pipeline(
pipeline_name="epidisco_3",
run_name=lambda patient: "epidisco_{}".format(patient.id),
config=EpidiscoConfig(cohort,
arg_normal_input=lambda patient: patient.normal_sample.bam_path_dna))
stomize the normal BAM input, the run name, and which patients to run on.
rt.add_epidisco_pipeline(
pipeline_name="epidisco_4",
run_name=lambda patient: "epidisco_{}".format(patient.id),
config=EpidiscoConfig(cohort,
keep=lambda patient: patient.id == "468",
arg_normal_input=lambda patient: patient.normal_sample.bam_path_dna))
me as #4, but written differently.
s EpidiscoConfigModified(EpidiscoConfig):
def keep(self, patient):
return patient.id == "468"
def arg_normal_input(self, patient):
return patient.tumor_sample.bam_path_dna
fied_config = EpidiscoConfigModified(cohort)
rt.add_epidisco_pipeline(
pipeline_name="epidisco_5",
config=modified_config)
stomize any CLI argument; for example, picard_java_max_heap_size.
rt.add_epidisco_pipeline(
pipeline_name="epidisco_6",
config=EpidiscoConfig(cohort,
arg_picard_java_max_heap_size="20g"))
is is Discohort's own dry run functionality, FYI. Should have a better name.
rt.run_pipeline("epidisco_1", dry_run=True)