VarCoPP predicts the pathogenicity of combinations of variants in gene pairs. It is based on digenic data present in DIDA, the Digenic Diseases Database, and it was trained against variants from the 1000 Genomes Project. VarCoPP consists of an ensemble of 500 individual Random Forest predictors and exploits variant, gene and gene pair features for its predictions.

At the moment, VarCoPP can be applied for Single Nucleotide Variants (SNVs) and small insertions/deletions from a single individual using the genome version GRCh37/hg19. It is highly recommended to perform beforehand an initial variant filtering procedure and generally restrict the analysis to variants from up to 150 genes.

VarCoPP was developed in the Interuniversity Institute of Bioinformatics in Brussels, under the collaboration of Université libre de Bruxelles and Vrije Universiteit Brussel.

Submit your variants

You can either insert each variant manually with the six column boxes, or copy-paste a complete variant list directly in the white box (tab or space delimited columns), or upload a VCF file. Please also provide gender information (if available). Further information on how to upload your data is provided in the "About" page.

Example for copy-pasted variant list:
1 69621 . A - Heterozygous
2 177054850 . C G Heterozygous
16 3254467 . CCTT C Heterozygous
X 107841975 . C A Homozygous

Example VCF file:


ULB_equalsize.001 kopie 2  (IB)2  vub.001