dbNSFP

A comprehensive, one-stop resource for functional annotations and deleteriousness predictions of all human non-synonymous and splice-site variants — trusted by researchers and clinical platforms worldwide since 2011.

85M+
non-synonymous & splice-site SNVs
36
prediction algorithms
2011
first release

Deleteriousness predictions

36 algorithms including CADD, REVEL, AlphaMissense, popEVE, ESM1b, MutPred2, and more — covering missense and splice-site variants genome-wide.

Population allele frequencies

gnomAD v4.1 & v2.1.1, TOPMed, All of Us, RGC Million Exome, 1000 Genomes, and ALFA — with population-stratified subsets.

Evolutionary conservation

PhyloP (3 versions), phastCons (3 versions), GERP++, GERP_92_mammals, and bStatistic across multi-species alignments.

T2T-CHM13 v2.0 support

v5.3.1 adds hs1 reference genome coordinates, fully supported in the web query service and annotation program.

ClinVar Benchmark

Interactive benchmark comparing all 36 pathogenicity prediction scores in dbNSFP against ClinVar variants (2017–2025 snapshots). Explore ROC curves, AUC metrics, and per-algorithm performance across ACMG genes — updated with every dbNSFP release.

View Benchmark →
Includes
  • ROC curves for all prediction scores
  • AUC & pAUC with 95% bootstrap CIs
  • ACMG secondary findings gene set
  • ClinVar snapshots 2017–2025

License

dbNSFP is free for academic and non-profit use. Commercial use requires an annual license. All users must comply with the terms of the original data sources (gnomAD, ClinVar, etc.).

Free

Academic & Non-Profit

  • Universities and research institutions
  • Government and public health agencies
  • Non-profit organizations
  • Personal and educational use
  • Publications and preprints

Commercial

  • Pharmaceutical & biotech companies
  • Clinical diagnostic laboratories
  • Software and SaaS platforms
  • Contract research organizations
  • Any for-profit application of dbNSFP
Citation requirement: All users (academic and commercial) must cite the appropriate dbNSFP publications when using the database in any publication or product. Please see www.dbnsfp.org/publications for the current citation list.

Questions? Contact us at license@dbnsfp.org.

About Genos

Genos Bioinformatics LLC was founded in 2024 in Texas, USA, by two computational genomic scientists dedicated to advancing human genetics and genomics. Our mission is to build sustainable, high-quality, and user-friendly data infrastructure that supports clinical genomic practice and genetics research, benefiting both the scientific community and the broader public.

Our flagship resource, dbNSFP, has been a cornerstone of human variant annotation and interpretation since 2011. It is widely used in academic research and serves as a foundational dataset for regulatory-compliant clinical genomics platforms worldwide.

Our Founders

Dr. Xiaoming Liu

Dr. Xiaoming Liu

Co-Founder

Dr. Xiaoming Liu is an Associate Professor at the University of South Florida College of Public Health. He holds a Ph.D. in Human Molecular Genetics from The University of Texas Graduate School of Biomedical Sciences, Houston, TX. His research focuses on developing statistical and computational methods for analyzing variants in human genomes from the perspectives of population genetics, bioinformatics, and genetic epidemiology. Dr. Liu created dbNSFP in 2011 and has since led its development into a comprehensive functional annotation database for human genetic variants.

Jin Yu

Jin Yu

Co-Founder

Jin Yu began his career in bioinformatics at the Human Genome Sequencing Center at Baylor College of Medicine in Houston in 2009, developing NGS algorithms for the 1000 Genomes Project. Since then he has worked across academia and the biotech industry, publishing 30+ papers in NGS, genomics, bioinformatics, and statistical genetics with over 40,000 citations. At Genos, he leads QA/QC, data architecture, and new product development for dbNSFP web services and AI applications.

Contact Us

License Inquiries

license@dbnsfp.org

Scientific & Business Collaboration

collaboration@dbnsfp.org

Feedback on dbNSFP & Related Services

feedback@dbnsfp.org