gatk4/gatk4_variantfiltration

variant filtering

GATK

GATK4

hard filtering

VCF

Description

Filter variant calls based on INFO and/or FORMAT annotations.

This tool is designed for hard-filtering variant calls based on certain
criteria. Records are hard-filtered by changing the value in the FILTER field
to something other than PASS. Filtered records will be preserved in the
output unless their removal is requested. Site-level filters are supplied as
one or more JEXL expressions via --filter_expression with a name via
--filter_name.

Type

bash_script

License

Apache-2.0

Keywords

variant filtering

GATK

GATK4

hard filtering

VCF

Links

Repository

Documentation

Contributors

Launch

Run this component

Run the following command to execute this component with Nextflow:

cat > params.yaml <<'EOM'  
output: "$id.$key.output.gz"  
id: "run"  
publish_dir: "output/"  
EOM

nextflow run https://packages.viash-hub.com/vsh/biobox.git \  
  -revision v0.5.0 \  
  -main-script target/nextflow/gatk4/gatk4_variantfiltration/main.nf \  
  -params-file params.yaml

GATK Engine Options

Name	Type & Properties	Description
--interval_padding -ip	integer	Amount of padding (in bp) to add to each interval specified via `--intervals`, to capture flanking context (e.g. near capture/exome-bait edges). Default: 0. A commonly-used value in GATK's own exome workflows is `100`.
--sites_only_vcf_output	boolean_true	Only emit site-level information in the output VCF/GVCF, dropping the FORMAT/genotype columns. Useful for producing a compact, privacy-preserving sites-only VCF (e.g. for public release or QC).
--create_output_variant_index -OVI	boolean	Create a VCF index (`.tbi`) when writing a coordinate-sorted VCF output. Default: true.
--create_output_bam_index -OBI	boolean	Create a BAM/CRAM index when writing a coordinate-sorted BAM/CRAM output. Default: true.
--output_cram_version	string	CRAM version to write for CRAM output ('3.0' or '3.1'). Ignored for BAM/SAM output. Note that CRAM 3.1 cannot be read by tools that only support CRAM 3.0. Default: '3.1'.
--read_filter -RF	string multiple	Read filter(s) to apply before analysis, given by their GATK Java class name (e.g. `MappingQualityReadFilter`). Can be specified multiple times.
--disable_read_filter -DF	string multiple	Read filter(s) to disable before analysis, given by their GATK Java class name. Can be specified multiple times.
--disable_tool_default_read_filters	boolean_true	Disable all tool-default read filters. Many tools will not function correctly without their default read filters enabled, so use with caution.
--disable_sequence_dictionary_validation	boolean_true	Do not check the input sequence dictionaries for compatibility. Useful when combining files from different sources with mismatched (but compatible) dictionaries. Use at your own risk.

Name

Type & Properties

--interval_padding

-ip

integer

--sites_only_vcf_output

boolean_true

--create_output_variant_index

-OVI

boolean

--create_output_bam_index

-OBI

boolean

--output_cram_version

string

--read_filter

-RF

string

multiple

--disable_read_filter

-DF

string

multiple

--disable_tool_default_read_filters

boolean_true

--disable_sequence_dictionary_validation

boolean_true

Input

Name	Type & Properties	Description
--variant -V	file required	Input VCF file containing variants to be filtered.
--reference -R	file	Reference genome FASTA file. Optional for `VariantFiltration` itself, but if provided, `--reference_fai` and `--reference_dict` must be provided too.
--reference_fai	file	FASTA index (`.fai`) file for the reference genome.
--reference_dict	file	Sequence dictionary (`.dict`) file for the reference genome.

Name

Type & Properties

--variant

-V

file

required

--reference

-R

file

--reference_fai

file

--reference_dict

file

Output

Name	Type & Properties	Description
--output -O	file required output	Output VCF file with the `FILTER` column annotated.

Name

Type & Properties

--output

-O

file

required

output

Options

Name	Type & Properties	Description
--apply_allele_specific_filters	boolean_true	Set the mask at the allele level rather than the site level. Not compatible with clustered-SNP filtering (`--cluster_size`/`--cluster_window_size`).
--cluster_size -cluster	integer	The number of SNPs which make up a cluster. Must be at least 2. Used together with `--cluster_window_size` to enable clustered-SNP filtering. Requires `--variant` to support random access (e.g. bgzipped and tabix-indexed, or a plain VCF with an accompanying `.idx`). Default: 3.
--cluster_window_size -window	integer	The window size (in bases) in which to evaluate clustered SNPs. A value of 0 disables clustered-SNP filtering. Requires `--variant` to support random access (e.g. bgzipped and tabix-indexed, or a plain VCF with an accompanying `.idx`). Default: 0.
--filter_expression -filter	string multiple	One or more JEXL expressions used with INFO fields to filter, e.g. `"vc.isSNP() && QD < 2.0"`. Must be paired 1-to-1 with `--filter_name`.
--filter_name	string multiple	Names for the list of site-level filters in `--filter_expression`. Recorded in the `FILTER` field for records that fail the corresponding expression.
--filter_not_in_mask	boolean_true	Filter records that are NOT in the regions given by `--mask`.
--genotype_filter_expression -G-filter	string multiple	One or more expressions used with FORMAT (sample/genotype-level) fields to filter, distinct from the site-level hard filters above. Must be paired 1-to-1 with `--genotype_filter_name`.
--genotype_filter_name -G-filter-name	string multiple	Names for the list of genotype-level filters (1-to-1 mapping with `--genotype_filter_expression`); recorded in the FORMAT `FT` field for genotypes that fail the corresponding expression.
--invalidate_previous_filters	boolean_true	Remove previous filters applied to the VCF before applying the filters requested here.
--invert_filter_expression -invfilter	boolean_true	Invert the selection criteria for `--filter_expression`.
--invert_genotype_filter_expression -invG-filter	boolean_true	Invert the selection criteria for `--genotype_filter_expression`.
--mask	file	BED/VCF file of genomic regions used to mask (flag) variants that fall within (or near, see `--mask_extension`) them.
--mask_description	string	Description to add to the FILTER field header line for the mask filter.
--mask_extension	integer	How many bases beyond the regions in `--mask` should variants also be filtered. Default: 0.
--mask_name	string	The text to put in the FILTER field if a variant overlaps (or is within `--mask_extension` of) the regions in `--mask`. Default: 'Mask'.
--missing_values_evaluate_as_failing	boolean_true	When evaluating the JEXL filter expressions, treat missing INFO/FORMAT values as failing the expression rather than passing it.
--set_filtered_genotype_to_no_call	boolean_true	Set genotypes that fail `--genotype_filter_expression` to no-call (`./.`) instead of leaving the original genotype call annotated with a FORMAT `FT` filter name.

Name

Type & Properties

--apply_allele_specific_filters

boolean_true

--cluster_size

-cluster

integer

--cluster_window_size

-window

integer

--filter_expression

-filter

string

multiple

--filter_name

string

multiple

--filter_not_in_mask

boolean_true

--genotype_filter_expression

-G-filter

string

multiple

--genotype_filter_name

-G-filter-name

string

multiple

--invalidate_previous_filters

boolean_true

--invert_filter_expression

-invfilter

boolean_true

--invert_genotype_filter_expression

-invG-filter

boolean_true

--mask

file

--mask_description

string

--mask_extension

integer

--mask_name

string

--missing_values_evaluate_as_failing

boolean_true

--set_filtered_genotype_to_no_call

boolean_true