Machine Learning-enabled Computational Discovery of Self-assembling Biocompatible Nanoaggregates

Electronically  active  organic  molecules  have  demonstrated  great promise as novel soft materials for energy harvesting and transport. Self-assembled   nanoaggregates   formed   from p-conjugated oligopeptides composed of an aromatic core flanked by oligopeptide wings offer emergent optoelectronic properties within a water-soluble and  biocompatible  substrate.  Deep  representational  learning  and Bayesian optimization were used to guide a course-grained molecular dynamics  simulation  to  effectively  screen  and  traverse  molecular space. The virtuous feedback between the data and domain science tools  enabled  identification  of  new  molecules  capable  of  self-assembling   biocompatible   nanoaggregates   with   emergent optoelectronic functionality. This work establishes new understanding of  oligopeptide  assembly,  identifies    promising  new  candidates  for experimental testing, and presents a computational design platform that can be generically extended to other peptide-based and peptide-like systems.

Additional Materials