Protein / RNA Folding by Learning

Abstract

Life is orchestrated via an interplay of many biomolecules. Any understanding of biomolecular function relies on detailed knowledge of their three-dimensional structure, whose determination is experimentally often very challenging. An orthogonal theoretical approach is the set of structure prediction techniques. The Helmholtz funded project ProFiLe aims to predict structures by using untapped information in the exponentially growing genomic databases via deep learning methods on high-performance computers. By this data-driven approach we want to accurately infer pairs of residues in spatial contact within biomolecules, as well as to guide structure prediction in a tailored open source software. ProFiLe employs transformer networks and their attention mechanism to learn in a self-supervised way the complex processes leading to protein or RNA folding.