GitHub

Genotypes Imputation

This workflow sends the genotypes for imputation to TOPMed, downloads and aggregates the results per chromosome. It follows the principles provided in their documentation.

Platform

At the moment, it is impossible to run this workflow from ODAP because the TOPMed servers have not been white listed. Instead you can run it from eddie.

Inputs

For this workflow to work, you will need:

  • Genotypes: likely the output of the Combining Datasets workflow.
  • A TOPMed token: see this page.
  • job ids (optional): Only to resume a crashed job.

Please have a look at the wokflow parameters below for how to setup the run.

Running The Workflow

If the previous steps have been completed successfully you can run:

nextflow run main.nf -entry Imputation -profile eddie -resume -with-report -with-trace -c run.config
Crash

The TOPMedImputation process waits for TOPMed imputation jobs to finish, which might be longer than the maximum eddie job duration (48h) depending on the server's queue size. In that case, the workflow will crash but all the TOPMed jobs should have been submitted. Resuming the workflow will thus not work and it will try to resubmit new jobs. In order to bypass this behaviour you can pass an optional TOPMED_JOBS_LIST to proceed directly to the download stage. These job ids can be obtained from the TOPMed urls.

Outputs

All outputs are produced in PUBLISH_DIR (defaults to results), the main outputs of the workflow are:

  • chr_P.qced.{pgen,pvar,psam}: A set of imputed genotypes, one for each chromosome in PGEN format.

Workflow Parameters

This is the list of all the pipeline's parameters, they can be set in the run.config file under the params section.

Inputs Parameters

  • GENOTYPES_PREFIX: Prefix to genotypes files.
  • TOPMED_TOKEN_FILE: Path to the file containing your TOPMed API token.

Important Options

  • TOPMED_ENCRYPTION_PASSWORD: An encryption password
  • TOPMED_JOBS_LIST: If the workflow crashes and you want to resume, list the job-ids in this file (one per line). Job ids can be obtained from the job url in TOPMed.
  • N_SAMPLES_PER_IMPUTATION_JOBS (default: 10000): We can only send file of less than 200000 samples to TOPMed and the server only allows 3 jobs at a time. This number ideally splits your data in 3 roughly equal batches.
  • IMPUTATION_R2_FILTER (default: 0.8): Only imputed variants passing the threshold are kept, set to 0 if you want to keep them all.

Secondary Options

  • TOPMED_REFRESH_RATE (default: 180): The frequency (in seconds) with which the workflow will monitor the imputation process to send further jobs.
  • TOPMED_MAX_PARALLEL_JOBS (default: 3): The maximum number of concurrent imputation processes, this is limited to 3 at the moment by TOPMed.