Another in our series of online industrial talks, this one is from Schrodinger.

Title: Accurate in silico prediction of protein thermostability with free energy perturbation

Speaker: Jianxin Duan

Abstract: Engineering thermostability is critical for the development of therapeutic proteins and industrial enzymes; not only does it influence protein activity at a certain temperature but is also associated with aggregation behaviour related to partial unfolding. Therefore, the ability to accurately predict the effect of mutations using computational methods represents a significant opportunity. In this presentation, we show free energy perturbation is able to predict relative thermostability changes of single point mutations with a mean-unsigned error of 0.86 kcal/mol across different mutation types. More importantly, we will discuss the current limitations and possible solutions.