Abstracts
New Machine Learning-Driven Approaches for Peptide Probe and Drug Discovery
Presenter: Arman Simonyan (gtk897@ku.dk)
Authors:
Arman Simonyan; Prof. Wouter Boomsma PhD; Prof. David Gloriam PhD
Copenhagen University
Two-thirds of human hormones act through ~800 G protein-coupled receptors (GPCRs). The vast majority (71%) of these hormones are peptides or proteins, which also account for an increasing share of drugs [1]. The study of peptide-receptor recognition is thus essential for understanding physiology, pathology and for drug design.
Our work aims to address the modeling problem of peptide-receptor recognition by leveraging AI-based methods and unique data from the field hub GPCRdb [2]. We benchmark the state-of-the-art in silico peptide design protocols, including RfDiffusion [3], AlfaFold Hallucination [4], ProteinMPNN [5], ESM-IF1 [6], Frame2seq [7], AFinitialGuess [8], on our dataset. In addition, we build in-house predictive statistical and deep leaning models representing residue interaction networks across the receptor-peptide interface. The models are trained on a unique data representation, storing data for individual residues rather than the overall protein. This allows peptide data to be inferred across conserved residues in different receptors – enabling use on receptors not targetable with classical methods.
The in silico discovered probes are being tested in vitro by pharmacological collaborators. In all, this will let us discover novel drugs and engineer new probes, enabling functional characterization of understudied receptors that cannot be targeted with current techniques.