FAQ

DNA Pipeline

How do I run variant calling on a single chromosome?

• Some tools allow you to specify a set of desired chromosomes within the dna_pipeline_config.yaml file; indicate the desired chromosome, and set seq_type: wgs
  • MuTect2, VarScan, GATK:CNV, NovoBreak (in development)
• For any other tool, the easiest approach would be to subset your BAMs to the desired chromosome and create a new data_config.yaml file. Again, be sure to set seq_type: wgs within the config file.

I want to annotate my variants (SNV and INDELs) with OncoKB - how do I do this?

• The OncoKB annotator requires internet access which means it can't be run directory on H4H. However, you can mount H4H and run it locally or transfer your mutation data to Samwise and run it there.
  • From H4H or Samwise, you can run the version of OncoKB located in /pughlab/bin.
  • Or, download the most recent version from git: oncokb-annotator
• Obtain an OncoKB access key: apiAccess

python3 /path/to/oncokb-annotator/MafAnnotator.py \
-t ONCOTREE_CODE \
-b OncoKB key \
-r reference \
-i /path/to/PROJECTNAME/Report/plots/date_PROJECTNAME_ensemble_mutation_data.tsv \
-o /path/to/PROJECTNAME/Report/plots/date_PROJECTNAME_ensemble_oncokb_annotated.tsv

You can then pass this data into filter_ensemble_mutations.R to perform filtering and update the SNV landscape plot.

module load R/3.6.1

Rscript /path/to/pipeline-suite/scripts/report/filter_ensemble_mutations.R \
-p PROJECTNAME \
-o /path/to/PROJECTNAME/Report/plots \
-t exome (or wgs) \
-i /path/to/PROJECTNAME/Report/plots/date_PROJECTNAME_ensemble_oncokb_annotated.tsv \
-c /path/to/PROJECTNAME/Report/data/total_bases_covered.tsv \
-m /path/to/PROJECTNAME/Report/data/msi_estimates.tsv