The group under Professor Tom Lenaerts (VUB-ULB) of the IB² has created an AI algorithm that facilitates the identification of combinations of genetic variants or abnormalities that lead to rare diseases through computer analysis.
The algorithm was created with Prof. Dr Guillaume Smits (Center for Human Genetics of the ULB, the Erasmus Hospital, and the University Children’s Hospital Queen Fabiola) and was designed and constructed in cooperation with Yves Moreau and Jan Aerts (KU Leuven), Sonia Van Dooren (UZ Brussel), and Ann Nowé (Vrije Universiteit Brussel). The technique has been given the name VarCoPP (Variant Combinations Pathogenicity Predictor).
Nearly 80% of rare diseases are genetically determined. It is thus important for doctors to be able to foresee which genetic variants in the patient’s genome may be the cause of the disease. It is not easy to predict the cause of the error but it is even more difficult to predict whether a combination of errors in different genes has the ability to cause a rare disease. However, this is essential for the better diagnoses of genetic diseases as, in several instances, only a small part of patients can be helped.
The VarCoPP algorithm exactly provides that innovative methodology: it allows simultaneous testing of the combinations of different variants in gene pairs and prediction of their potential pathogenicity. The AI that lies beneath VarCoPP is driven by the database of rare diseases known as DIDA (dida.ibsquare.be), which was created by the same scientists in 2015. The scientists have succeeded in testing the effectiveness and reliability of the algorithm on 23 independent pathogenic gene combinations, and offer confidence intervals of 95% and 99% to assist doctors to zoom in on the most vital predictions.
The team is currently trying to use these results to find the genetic causes of rare diseases in patients for whom no cause could earlier be identified. The group simultaneously launches a new online diagnostic platform for scientists and clinicians, based on the algorithm. The platform is known as “ORVAL” and is reported in a publication in the journal Nucleic Acids Research (NAR).
ORVAL and VarCoPP offer a new methodology to investigate variant combinations for rare diseases for which causal genes are either known or unknown, such as the hundreds of autism or epilepsy genes or the 20 genes of the rare Bardet-Biedl syndrome (a genetic disorder which causes obesity, blindness, and motor disorders among others) where different combinations of genetic changes are the possible cause.