Integrating AI, automation and multiscale simulations for end-to-end design of phase-separating proteins

Arvind Ramanathan

Abstract:

Liquid-liquid phase separation (LLPS) is a fundamental cellular process that is
driven by self-assembly of intrinsically disordered proteins (IDPs), protein-RNA
complexes, or other bio-molecular systems which can form liquid droplets. Many
natural materials including silk, elastin, and gels are a result of LLPS and thus
rational design of such phase-separating peptides can have transformative impact, from designing new biologically inspired materials (e.g., clothing) to selfcompartmentalized drug-delivery systems for biomedical applications. However,
given the intrisinc complexity in the rules governing LLPS, rational design of LLPS
undergoing peptides remains challenging. We posit that automation, foundation
models integrated with reinforcement learning approaches and multiscale molecular
simulations can drive the design of novel peptides that undergo LLPS. We describe
our progress towards the goal of end-to-end design of phase separating peptides
by summarizing current work at the Argonne National Laboratory’s Advanced
Photon Source 8ID-I beamline, where a robotic set up in the laboratory is enabled
via simulation and extensive testing of such bio-materials. Together, our approach
enables the design of novel bio-materials that can undergo phase separation under
diverse physiological conditions

Paper Track: Papers

Submission Category: AI-Guided Design