What is dbNSFP? 

dbNSFP is a comprehensive, one-stop resource for functional annotations and deleteriousness predictions of all possible non-synonymous single-nucleotide variants (nsSNVs) in the human genome. Since its initial release in 2011, dbNSFP has established itself as a trusted data infrastructure advancing human genetic and clinical genomic research (Publications) and powering regulatory-compliant clinical genomics platforms worldwide.

Its current version is based on GENCODE Human release 48 (Ensembl version 114) and includes a total of 81.5M nsSNVs and 2.2M ssSNVs (splicing-site SNVs) derived from all known protein-coding genes in the human genome (includes mitochondrial DNA).

dbNSFP compiles functional annotations and deleteriousness prediction scores from 34 algorithms: 

• SIFT, SIFT4G, PROVEAN, Polyphen2-HDIV, Polyphen2-HVAR, MutationTaster 2021, MutationAssessor, FATHMM-XF coding, CADD, VEST4, DANN, MetaSVM, MetaLR, MetaRNN, Eigen, Eigen-PC, M-CAP, REVEL, MutPred2, MVP, gMVP, MPC, PrimateAI, DEOGEN2, ALoFT, BayesDel, ClinPred, LIST-S2, VARITY, ESM1b, AlphaMissense, PHACTboost, MutFormer, and MutScore.

Evolutionary conservation scores:

• PhyloP (3 versions), phastCons (3 versions), GERP++, GERP_91_mammals, and bStatistic.

and observed allele frequencies from large population sequencing projects:

• The 1000 Genomes Project, gnomAD v4.1 and v2.1.1 (including non-neuro, non-cancer, and control sample subsets), TOPMed, All of Us, RGC Million Exome, and ALFA (aggregated from dbGaP and dbSNP)

Moreover, dbNSFP curates comprehensive gene-level annotations, including:

• Various gene IDs from HGNC

• Gene-Disease Relationships: GenCC, OMIM, Orphanet, The Human Phenotype Ontology, GWAS Catalog, ClinGen Dosage Sensitivity, etc.

• Protein functions and structures: The Human Protein Atlas, UniProt, Gene Ontology, IntAct, etc.

• Metabolic and signaling pathways: ConsensusPathDB, KEGG pathway, etc.

• Model organism experiment data: MGI, ZFIN, etc.

For a full list of data sources and version information, please refer to the current README of dbNSFP in the Releases page.

Developer Collaboration

We welcome developers of functional prediction methods to provide their predictions and scores to the database. Please contact us at collaboration@dbnsfp.org.