Efficient Machine Learning Pipeline Predicts Nanoscale Interactions Location

A new machine learning tool called NeCLAS, developed by researchers at the University of Michigan, can predict how proteins and nanoparticles might interact, including predicting binding sites and the likelihood of binding between them. This new tool could aid in the fight against antibiotic-resistant infections and new viruses, by designing drugs that target crucial proteins in bacteria and viruses without harming human cells. NeCLAS uses structural models of proteins and their known interaction sites to extrapolate how proteins and nanoparticles might interact and predict interactions between two proteins or two nanoparticles. This enables researchers to understand the potential applications of nanoparticles and optimize their designs. The tool was tested with three case studies, showing the potential of the tool in optimizing designs and predicting interactions. The study was funded by the Army Research Office, National Science Foundation and the University of Michigan Blue Sky Initiative.