AutoGVP: Automated Germline Variant Pathogenicity

This work has now been published: AutoGVP: a dockerized workflow integrating ClinVar and InterVar germline sequence variant classification.

Kim J^, Naqvi AS^, Corbett RJ, Kaufman RS, Vaksman Z, Brown MA, Miller DP, Phul S, Geng Z, Storm PB, Resnick AC, Stewart DR, Rokita JL+, Diskin SJ+. AutoGVP: a dockerized workflow integrating ClinVar and InterVar germline sequence variant classification. Bioinformatics. 2024 Mar 4;40(3):btae114. doi: 10.1093/bioinformatics/btae114. PMID: 38426335; PMCID: PMC10955249.

^Equal first authorship +Equal senior authorship

AutoGVP Workflow

For more detailed instructions, please visit the user guide on our wiki.

Clone the AutoGVP repository

git clone git@github.com:diskin-lab-chop/AutoGVP.git

Docker set-up

1. Pull the docker image.

docker pull pgc-images.sbgenomics.com/diskin-lab/autogvp:v2.0.1

2. Navigate to the AutoGVP root directory

cd AutoGVP

3. Start a docker image. Replace <CONTAINER_NAME> with any name and run the commands below:

docker run --platform linux/amd64 --name <CONTAINER_NAME> -d -v $PWD:/home/rstudio/AutoGVP pgc-images.sbgenomics.com/diskin-lab/autogvp:v2.0.1
docker exec -ti <CONTAINER_NAME> bash

4. Navigate to AutoGVP directory within the docker image

cd /home/rstudio/AutoGVP

5. Run AutoGVP (see example commands below).

Dependencies

VEP (v104)
InterVar
ANNOVAR
AutoPVS1 (v2.0)
bcftools (v1.17)

How to Run AutoGVP

AutoGVP Requirements (recommended to place all in the data/ folder):

• VEP-annotated VCF file (*VEP.vcf)
• ANNOVAR multianno file (*hg38_multianno.txt)
• InterVar file (*intervar.hg38_multianno.txt.intervar)
• AutoPVS1 file (*autopvs1.txt)
• ClinVar variant summary file (variant_summary_YYYY-MM.txt.gz)
• ClinVar submission summary file (submission_summary_YYYY-MM.txt.gz)
• Variant submissions file (resolved-clinvar-YYYY-MM[-concept-conflict_res].tsv generated by resolve-clinvar-intepretations.R)
• ClinVar hgvs4variation file (if running update_intervar.R, which is executed by default)

Custom workflow example run

1. Prepare input files by running VEP, ANNOVAR, InterVar, and AutoPVS1.
2. Download database files:

bash scripts/download_db_files.sh

3. Run resolve-clinvar-intepretations.R. To customize conflicting interpretation resolution, users can provide a ClinGen Concept ID list to filter submissions against (--conceptID_list). When a list is provided, users can also determine how unsettled conflicts are resolved with the --conflict_res argument ("latest" or "most_severe"). For more details, see the FAQ. Example command:

Rscript scripts/resolve-clinvar-intepretations.R --variant_summary data/variant_summary_YYYY-MM.txt.gz --submission_summary data/submission_summary_YYYY-MM.txt.gz --outdir refs --conceptID_list refs/clinvar_<concept_cat>_concept_ids_YYYYMMDD.txt --conflict_res "latest"

4. Run AutoGVP; if output of scripts/resolve-clinvar-intepretations.R is not provided, the script will be run prior to starting pathogenicity assessment

bash run_autogvp.sh \
--vcf=data/test_VEP.vcf \
--filter_criteria=<filter criteria> \
--intervar=data/test_VEP.hg38_multianno.txt.intervar \
--multianno=data/test_VEP.vcf.hg38_multianno.txt \
--autopvs1=data/test_autopvs1.txt \
--hgvs4variation_file=data/hgvsv4variation-2026-07.txt.gz \
--outdir=results \
--out="test_custom" \
--selected_clinvar_submissions=refs/resolved-clinvar-YYYY-MM.tsv \
--variant_summary=data/variant_summary_YYYY-MM.txt.gz \
--submission_summary=data/submission_summary_YYYY-MM.txt.gz \
--conceptIDs=refs/clinvar_<concept_cat>_concept_ids_YYYYMMDD.txt \
--conflict_res="latest"

CAVATICA workflow example run

1. Download database files:

bash scripts/download_db_files.sh

2. Run resolve-clinvar-intepretations.R (See custom workflow step 3 for optional conflict resolution parameters). For more details, see the FAQ. Example command:

Rscript scripts/resolve-clinvar-intepretations.R --variant_summary data/variant_summary_YYYY-MM.txt.gz --submission_summary data/submission_summary_YYYY-MM.txt.gz --outdir refs --conceptID_list refs/clinvar_<concept_cat>_concept_ids_YYYYMMDD.txt --conflict_res "latest"

3. Run AutoGVP; if output of scripts/resolve-clinvar-intepretations.R is not provided, the script will be run prior to starting pathogenicity assessment

bash run_autogvp.sh \
--vcf=data/test_pbta.single.vqsr.filtered.vep_105.vcf \
--filter_criteria=<filter criteria> \
--intervar=data/test_pbta.hg38_multianno.txt.intervar \
--multianno=data/test_pbta.hg38_multianno.txt \
--autopvs1=data/test_pbta.autopvs1.tsv \
--hgvs4variation_file=data/hgvsv4variation-2026-07.txt.gz \
--outdir=results \
--out="test_pbta" \
--selected_clinvar_submissions=refs/resolved-clinvar-YYYY-MM.tsv \
--variant_summary=data/variant_summary_YYYY-MM.txt.gz \
--submission_summary=data/submission_summary_YYYY-MM.txt.gz \
--conceptIDs=refs/clinvar_<concept_cat>_concept_ids_YYYYMMDD.txt \
--conflict_res="latest"

AutoGVP Output

AutoGVP produces an abridged output file with minimal information needed to interpret variant pathogenicity, as well as a full output with >100 variant annotation columns.

Abridged output example:

chr	start	ref	alt	rs_id	gene_symbol_vep	variant_classification_vep	HGVSg	HGVSc	HGVSp	autogvp_call	autogvp_call_reason	clinvar_stars	clinvar_clinsig	intervar_evidence
chr1	1332490	C	T	rs201607183	TAS1R3	missense_variant	chr1:g.1332490C>T	c.959C>T	p.Thr320Met	Uncertain_significance	ClinVar	1	Uncertain_significance	InterVar: Uncertain significance PVS1=0 PS=[0, 0, 0, 0, 0] PM=[1, 0, 0, 0, 0, 0, 0] PP=[0, 0, 1, 0, 0, 0] BA1=0 BS=[0, 0, 0, 0, 0] BP=[0, 0, 0, 0, 0, 0, 0, 0]
chr1	1390349	C	T	rs769726291	CCNL2	missense_variant	chr1:g.1390349C>T	c.887G>A	p.Gly296Asp	Uncertain_significance	InterVar	NA	NA	InterVar: Uncertain significance PVS1=0 PS=[0, 0, 0, 0, 0] PM=[1, 1, 0, 0, 0, 0, 0] PP=[0, 0, 0, 0, 0, 0] BA1=0 BS=[0, 0, 0, 0, 0] BP=[0, 0, 0, 0, 0, 0, 0, 0]

See here for list of columns included in full output. Users can provide their own file specifying column names to include in final output with the --output_colnames argument (please use same formatting as output_colnames_default.tsv).

Code Authors

Ammar S. Naqvi (@naqvia), Ryan J. Corbett (@rjcorb), and Patricia J. Sullivan (@pj-sullivan).

Contact

For questions, please submit an issue