Abstract

Information processing in the brain relies on signal transmission at neuronal synapses. Neurotransmitter release and generation of action potentials depend on protein-protein interactions (PPIs) networks. Genetic variants that alter the synaptic PPIs result in neurological and neuropsychiatric diseases, such as schizophrenia and autism spectrum disorders. To gain information on the effects of such variants we are predicting the effect of disease-associated protein variants in the synaptic interactome by using a combination of machine learning and bioinformatics. Many disease-associated mutations are located at protein-protein interfaces, thus causing disrupted or erroneous protein interactions. The so-called hot-spot residues (i.e. interfacial residues that contribute the most to protein-protein binding) are often enriched in disease-associated mutations. Thus, identifying if disease-associated variants corresponds to such hot spots can improve our understanding of their effects at the molecular level. In this direction, we are setting up an automatic pipeline for the prediction of hot-spots on the PPIN of the human synaptosome.

Host: Ira Assent