Common mutations in cancer that will be genotyped and passed through into the merged VCF if they have even low-level evidence of a mutation (by default, marked with filter DOCM_ONLY)

reference contains the nucleotide sequence for a given assembly (hg37, hg38, etc.) in fasta format for the entire genome. This is what reads will be aligned to. Appropriate files can be found on ensembl at https://ensembl.org/info/data/ftp/index.html When providing the reference secondary files corresponding to reference indices must be located in the same directory as the reference itself. These files can be created with samtools index, bwa index, and picard CreateSequenceDictionary.

tumor_name provides a string for what the MT sample will be referred to in the various outputs, for example the VCF files.

netmhc_stab sets an option that decides whether it will run NetMHCStabPan after all filtering and add stability predictions to predicted epitopes.

normal_name provides a string for what the WT sample will be referred to in the various outputs, for example the VCF files.

scatters each supported variant detector (varscan, pindel, mutect) into this many parallel jobs

bait_intervals is an interval_list corresponding to the baits used in sequencing reagent. These are essentially coordinates for regions you were able to design probes for in the reagent. Typically the reagent provider has this information available in bed format and it can be converted to an interval_list with Picard BedToIntervalList. AstraZeneca also maintains a repo of baits for common sequencing reagents available at https://github.com/AstraZeneca-NGS/reference_data

bqsr_intervals provides an array of genomic intervals for which to apply GATK base quality score recalibrations. Typically intervals are given for the entire chromosome (chr1, chr2, etc.), these names should match the format in the reference file.

tumor_sequence represents the sequencing data for the MT sample as either FASTQs or BAMs with accompanying readgroup information. Note that in the @RG field ID and SM are required.

net_chop_method is used to specify which NetChop prediction method to use (\"cterm\" for C term 3.0, \"20s\" for 20S 3.0). C-term 3.0 is trained with publicly available MHC class I ligands and the authors believe that is performs best in predicting the boundaries of CTL epitopes. 20S is trained with in vitro degradation data.

normal_sequence represents the sequencing data for the WT sample as either FASTQs or BAMs with accompanying readgroup information. Note that in the @RG field ID and SM are required.

pvacseq_threads specifies the number of threads to use for parallelizing peptide-MHC binding prediction calls.

Known polymorphic indels recommended by GATK for a variety of tools including the BaseRecalibrator. This is part of the GATK resource bundle available at http://www.broadinstitute.org/gatk/guide/article?id=1213 File should be in vcf format, and tabix indexed.

target_intervals is an interval_list corresponding to the targets for the capture reagent. BED files with this information can be converted to interval_lists with Picard BedToIntervalList. In general for a WES exome reagent bait_intervals and target_intervals are the same.

tumor_sample_name is the name of the tumor sample being processed. When processing a multi-sample VCF the sample name must be a sample ID in the input VCF #CHROM header line.

net_chop_threshold specifies the threshold to use for NetChop prediction; increasing the threshold results in better specificity, but worse sensitivity.

normal_sample_name is the name of the normal sample to use for phasing of germline variants.

An optional VCF with variants that will be flagged as 'VALIDATED' if found in this pipeline's main output VCF

ensembl species - Must be present in the cache directory. Examples: homo_sapiens or mus_musculus

ensembl version - Must be present in the cache directory. Example: 95

Determines whether variants found only via genotyping of DOCM sites will be filtered (as DOCM_ONLY) or passed through as variant calls

genome assembly to use in vep. Examples: GRCh38 or GRCm38

custom type, check types directory for input format

The effective coverage of capture products generally extends out beyond the actual regions targeted. This parameter allows variants to be called in these wingspan regions, extending this many base pairs from each side of the target regions.

used to provide clinical HLA typing results in the format HLA-X*01:02[/HLA-X...] when available.

used to provide clinical HLA typing results; separated from class I due to nomenclature inconsistencies

run_reference_proteome_similarity sets an option that decides whether it will run reference proteome similarity after all filtering and BLAST peptide sequences against the reference proteome to see if they appear elsewhere in the proteome.