Imene Boudellioua obtained her Computer Bachelor Science degree from Sultan Qaboos University, Oman and her Master’s Degree from King Abdullah University of Science and technology. She is currently a Ph.D. in King Abdullah University of Science and Technology and works in the Bio-Onthology Research Group under the supervision of Professor Robert Hoehndorf. Her research interests involve the application of machine learning and data mining algorithms for functional annotation of various biological entities including the prediction of proteins’ functions and human disease-causative genetic variation.
________________________________________________________________________________________________________
Discriminating the causative disease variant(s) for individuals with inherited or de novo mutations presents one of the main challenges the clinical genetics community faces today. Computational approaches for variant prioritization include machine learning methods utilizing a large number of features, including molecular information, interaction networks, or phenotypes. Here, we demonstrate the PhenomeNET Variant Predictor (PVP) system that exploits semantic technologies and automated reasoning over genotype-phenotype relations to filter and prioritize variants in whole exome and whole genome sequencing datasets. We demonstrate the performance of PVP in identifying causative variants on a large number of synthetic whole exome and whole genome sequences, covering a wide range of diseases and syndromes. In a retrospective study, we further illustrate the application of PVP for the interpretation of whole exome sequencing data in patients suffering from congenital hypothyroidism. We find that PVP accurately identifies causative variants in whole exome and whole genome sequencing datasets and provides a powerful resource for the discovery of causal variants.