The team of Tom Lenaerts, of the ULB-VUB Interuniversity Institute of Bioinformatics in Brussels (IB²), has developed an algorithm of artificial intelligence that can identify the potential genetic causes of rare diseases from a computer analysis.
80% of rare diseases are genetic, but predicting that such a variant in a patient's genome is the cause of a disease remains a complex task for clinicians. Even more if it is a combination of variants in different genes . For the first time, this algorithm - called VarCoPP (for Variant Combinations Pathogenicity Predictor ) - predicts the impact of combinations of genetic variations on the development of rare diseases. It makes it possible to test, in one go, the combinations of numerous variants in pairs of genes and to predict their potential pathogenicity.
In addition to Tom Lenaerts' team (ULB-VUB Interuniversity Institute of Bioinformatics in Brussels, IB²) , the agorithm was designed in co-promotion with Guillaume Smits (ULB's Center for Human Genetics, Erasme Hospital and University Hospital Des Queen Fabiola Children) and in collaboration with researchers from VUB, KU Leuven and UZ Brussel. It is described in an article of PNAS magazine .
It is also available online to provide a new online diagnostic tool for clinicians. The platform called "ORVAL" is described in a publication in the journal Nucleid Acids Research . VarCoPP could investigate and provide suggestions for diagnostic combinations for any disease for which causal genes are unknown or known such as, for example, the hundreds of autism or epilepsy genes or the Bardet-Biedl Malformation Syndrome genes.
provided by Université libre de Bruxelles.