> For the complete documentation index, see [llms.txt](https://academy.dnanexus.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://academy.dnanexus.com/public-datasets-on-the-dnanexus-platform/validation-ready-accelerator/gnomad-sites-only-for-dna-variant-filtering-in-clinical-pipelines.md). # gnomAD (sites only) – for DNA variant filtering in clinical pipelines ## Necessary Disclaimers and Legal The user is responsible for reviewing and complying with the license requirements of the software, and data referenced in this documentation. Users are responsible for the costs associated with analyzing data, software and its storage in their project spaces. Instance type availability and pricing are subject to the contract between the user or the user’s organization and DNAnexus. ## Citation for gnomAD Official introduction: Stat page: List of publication: Release blog: ## Overview The Genome Aggregation Database (gnomAD) was originally launched as ExAC in 2014. It is a global collaborative effort that aggregates and harmonizes large-scale human exome and genome sequencing data. Its goal is to provide high-quality variant frequency data to the scientific community. All datasets are reprocessed using standardized pipelines to ensure consistency. The latest release, gnomAD v4.1, includes over 730,000 exomes and 76,000 genomes (GRCh38). It introduces several key improvements, including fixing allele number issues and reporting allele numbers across all callable sites. It also integrates joint allele frequencies from both exomes and genomes, and flags variants with highly discordant frequencies between the two datasets. gnomAD v4.1 also introduces the GroupMax Filtering Allele Frequency metric. This is an ancestry-specific frequency measure. It is recommended by ACMG/ClinGen guidelines for more accurate classification of benign variants. However, its values may vary across versions as population diversity expands. ### Statistics * v4 release is composed of 730,947 exomes and 76,215 genomes (GRCh38). * gnomAD v4 structural variants (SV) represent 63,046 genomes (GRCh38). * gnomAD v4 copy number variants (CNV) represent variants in less than 1% of 464,297 exomes (GRCh38). ## gnomAD data on DNAnexus platform We provide gnomAD v4.1 variant datasets sourced from [the official repository](https://gnomad.broadinstitute.org/downloads#v4). Please access data [here in the platform](https://platform.dnanexus.com/panx/projects/J3JyY6j030gzQypGpk273241/data/gnomad_v4). The following datasets are available on our platform: * Exomes (VCF format). * Genomes (VCF format). * Joint Frequency data (Exomes + Genomes). Although gnomAD distributes both Hail Tables and VCFs, we specifically downloaded and hosted VCF files split by chromosome (chr1–22, X, Y) for compatibility and ease of use. Each directory contains per-chromosome VCF files (chr1–22, X, Y) and corresponding index files (.tbi). ``` gnomad_v4/ ├── exomes/ │ ├── gnomad.exomes.v4.1.sites.chr1.vcf.bgz │ ├── gnomad.exomes.v4.1.sites.chr1.vcf.bgz.tbi │ ├── gnomad.exomes.v4.1.sites.chr2.vcf.bgz │ ├── gnomad.exomes.v4.1.sites.chr2.vcf.bgz.tbi │ ├── ... │ ├── gnomad.exomes.v4.1.sites.chrX.vcf.bgz │ ├── gnomad.exomes.v4.1.sites.chrX.vcf.bgz.tbi │ └── gnomad.exomes.v4.1.sites.chrY.vcf.bgz │ ├── genomes/ │ ├── gnomad.genomes.v4.1.sites.chr1.vcf.bgz │ ├── gnomad.genomes.v4.1.sites.chr1.vcf.bgz.tbi │ ├── ... │ └── gnomad.genomes.v4.1.sites.chrY.vcf.bgz │ └── joint_frequency/ ├── gnomad.joint.v4.1.sites.chr1.vcf.bgz ├── gnomad.joint.v4.1.sites.chr1.vcf.bgz.tbi ├── ... └── gnomad.joint.v4.1.sites.chrY.vcf.bgz ``` ## How to use gnomAD data in nf-core/sarek pipeline ### Variant annotation in nf-core/sarek: The nf-core/sarek workflow uses the following tools for variant annotation (): * snpEff * snpSift (available since version 3.8.0 Feb 10) * bcftools annotate * VEP In the nf-core/sarek pipeline, gnomAD can be utilized in three annotation tools except snpEff. Note: The gnomAD VCF files are distributed in .bgz format, which is functionally identical to .gz but does not pass nf-core/sarek's input validation, which strictly requires files to end with .vcf or .vcf.gz. This can be resolved by simply renaming the files from .vcf.bgz to .vcf.gz without any file conversion, as the file content remains unchanged. ### VEP Cache v115 The Ensembl Variant Effect Predictor (VEP) cache is a pre-built local repository that enables high-throughput, offline variant annotation. Sarek pipeline uses VEP cache for variant annotation and allows users to specify their interested cache version via the --vep\_cache\_version parameter, with nf-core/sarek automatically pulling the corresponding cache from the nf-core S3 bucket s3://annotation-cache/vep\_cache/ if no local cache path is provided. It is essential to ensure that the VEP tool version (--vep\_version) and the cache version (--vep\_cache\_version) are identical. Please refer to the VEP[ cache blog](https://www.ensembl.info/2020/10/23/cool-stuff-ensembl-vep-can-do-whats-in-the-cache-and-how-does-vep-use-it/) and [official document](https://asia.ensembl.org/info/docs/tools/vep/script/vep_cache.html?#cache) for more information about VEP cache. The latest nf-core/sarek pipeline (version 3.8.1) runs on VEP Cache v115, which compiles ClinVar (Feb 2025), gnomAD v4.1 (Exomes/Genomes), dbSNP 156, and COSMIC 101 into a single, high-performance resource. It also features GENCODE 49 gene models and functional scores from SIFT and PolyPhen-2. Detailed metadata regarding these datasets and population-specific frequencies can be found in the provided info.txt file ``` # VEP Cache 115 Configuration Summary (from info.txt file) Species: Homo sapiens (GRCh38.p14) Annotations: - Gene_Models: GENCODE 49 - Variant_Frequencies: - gnomAD_v4.1 (Exomes & Genomes) - 1000Genomes_Phase3 - Clinical_Databases: - ClinVar_202502 - COSMIC_101 - dbSNP_156 - HGMD-PUBLIC_20204 - Impact_Scores: - SIFT_6.2.1 - PolyPhen_2.2.3 - Regulation: Ensembl_RegBuild_1.0 ``` In addition to the gnomAD data folder, we provide the VEP Cache v115 folder, which is compatible with the nf-core/sarek pipeline on DNAnexus. Users can utilize this resource by specifying the directory path in the --vep\_cache parameter as shown in the example below. ``` vep_cache dx://project-xxx:/sarek_dataset/vep_cache/ ``` ### SnpSift SnpSift annotates VCF files with custom annotation databases (e.g., dbSNP, gnomAD, ClinVar). To enable it, add “snpsift” to --tools and provide a CSV samplesheet via --snpsift\_databases. The CSV samplesheet specifies annotation databases. See [assets/snpsift\_databases\_example.csv](https://github.com/nf-core/sarek/blob/master/assets/snpsift_databases_example.csv) Here is our example: ```` ```csv vcf,tbi,fields,prefix,vardb /path/to/clinvar.vcf.gz,/path/to/clinvar.vcf.gz.tbi,ALLELEID;CLNDN;ORIGIN;CLNVC;CLNSIGCONF,ClinVar_, /path/to/gnomad.exomes.v4.1.sites.chr2.vcf.gz,/path/to/gnomad.exomes.v4.1.sites.chr2.vcf.gz.tbi,AF;AF_afr;AF_amr;cadd_phred;sift_max;revel_max,gnomAD_, /path/to/gnomad.exomes.v4.1.sites.chr21.vcf.gz,/path/to/gnomad.exomes.v4.1.sites.chr21.vcf.gz.tbi,AF;AF_afr;AF_amr;cadd_phred;sift_max;revel_max,gnomAD_, ``` ```` In the above example, replace /path/to/ with the actual directory where your database files are stored, such as clinvar.vcf.gz and the gnomAD per-chromosome VCF files. Due to large file sizes, we provided an example of a configuration file to increase the memory (to 16G) to avoid out of memory issues. ``` process { withName: 'SNPSIFT_ANNMEMCREATE' { ext.args = "-Xmx16G" } withName: 'SNPSIFT_ANNMEM' { ext.args = "-Xmx32G" } } ``` ### bcftools annotate To annotate variants with gnomAD using BCFtools in Sarek, you must provide four parameters: --bcftools\_annotations and --bcftools\_annotations\_tbi for the ClinVar VCF and its index; --bcftools\_columns for a file specifying the INFO fields to extract (e.g., AF, cadd\_phred,sift\_max); and --bcftools\_header\_lines for VCF header definitions. Although not required by BCFtools itself, these header lines are essential for nf-core/sarek's input validation. For --bcftools\_columns, users can follow this example config files to annotate using all available fields in the INFO column. ``` process { withName: 'NFCORE_SAREK:SAREK:VCF_ANNOTATE_ALL:BCFTOOLS_ANNOTATE' { ext.args = { '--columns INFO --output-type z --write-index=tbi' } } } ```