REVEL is an ensemble method for predicting the pathogenicity of missense variants based on a combination of scores from 13 individual tools: MutPred, FATHMM v2.3, VEST 3.0, Polyphen-2, SIFT, PROVEAN, MutationAssessor, MutationTaster, LRT, GERP++, SiPhy, phyloP, and phastCons.  REVEL was trained using recently discovered pathogenic and rare neutral missense variants, excluding those previously used to train its constituent tools.  The REVEL score for an individual missense variant can range from 0 to 1, with higher scores reflecting greater likelihood that the variant is disease-causing.  When applied to two independent test sets, REVEL had the best overall performance (p<10-12) compared with any individual tool and seven ensemble methods: MetaSVM, MetaLR, KGGSeq, Condel, CADD, DANN, and Eigen.  Importantly, REVEL also had the best performance for distinguishing pathogenic from rare neutral variants with allele frequencies <0.5%.  Compared with other ensemble methods, the area under the receiver operating characteristic curve (AUC) for REVEL was 0.046-0.182 higher in an independent test set of 935 recent SwissVar disease variants and 123,935 putatively neutral exome sequencing variants, and 0.027-0.143 higher in an independent test set of 1953 pathogenic and 2406 benign variants recently reported in ClinVar.  We provide precomputed REVEL scores for all possible human missense variants to facilitate the identification of pathogenic variants in the sea of rare variants discovered as sequencing studies expand in scale.

Histogram of REVEL scores 

Figure. REVEL score distributions for 6182 disease mutations from HGMD, 123,706 putatively neutral variants, and 1,125,160 exome sequence variants (ESVs) reported by the Exome Sequencing Project, Atherosclerosis Risk In Communities, and 1000 Genomes Project (left).  Sensitivity (true positive rate or recall) and specificity (true negative rate) for all possible REVEL score thresholds (right).

To aid in interpretation of REVEL scores, tables of sensitivity, specificity, LR+, and LR- values can viewed online or downloaded as a CSV or Excel file.


Ioannidis NM,* Rothstein JH,* Pejaver V, Middha S, McDonnell SK, Baheti S, Musolf A, Li Q, Holzinger E, Karyadi D, Cannon-Albright LA, Teerlink CC, Stanford JL, Isaacs WB, Xu J, Cooney KA, Lange EM, Schleutker J, Carpten JD, Powell IJ, Cussenot O, Cancel-Tassin G, Giles GG, MacInnis RJ, Maier C, Hsieh CL, Wiklund F, Catalona WJ, Foulkes WD, Mandal D, Eeles RA, Kote-Jarai Z, Bustamante CD, Schaid DJ, Hastie T, Ostrander EA, Bailey-Wilson JE, Radivojac P, Thibodeau SN, Whittemore AS, and Sieh W.  “REVEL: An ensemble method for predicting the pathogenicity of rare missense variants.”  American Journal of Human Genetics 2016; 99(4):877-885. http://dx.doi.org/10.1016/j.ajhg.2016.08.016

*These authors contributed equally and are listed alphabetically