BACKGROUND

Whole Exome Sequencing (WES) is increasingly applied to clinical diagnosis and shows about 50 to 90,000 variations per individual. Among these, one or two are pathogenic mutations responsible for Mendelian genetic diseases.

It is crucial to filter out non-pathogenic variants and limit downstream analysis to a few candidate mutations.

The UMD-Predictor system allows to efficiently annotate coding variations for their potential pathogenicity.

This system uses a combinatorial approach that is now available for all exonic nucleotide substitutions of the human genome.

KEY BENEFITS vs. STATE OF THE ART

 •  UMD-Predictor algorithm is the most efficient tool to predict pathogenic mutations with a positive predictive value of 99.4%, a sensitivity of 95.4%, and a specificity of 92.2%

•  Original combinatorial approach

•  Faster response: 180 s for a VCF file, down to 45 s with webservice

•  Better Diagnosis Odd Ratio (2 logs difference) compared to the best in others systems (SIFT, PolyPhen 2, Provean, Mutation Assessor, Condel, MutationTaster and CADD)

•  Direct consultation of an online database that stores all potential nucleotide substitution (≈268 million registrations): easily and quickly accessible for users

DEVELOPMENT STATUS

•  Existing website

•  Proven efficiency for mass data processing

•  No more development required

APPLICATIONS

Genome decryption

High-throughput DNA sequencing and genetic analysis

Clinical diagnostic: Predicting risk of developing diseases