Finding protein thermostability and spin-coupling constant using Bayesian statistics

摘要

This paper outlines the use of Bayesian statistics to find the thermostability and spin-coupling constant of protein. Thermostability is an important factor in protein efficacy; modeling it lets us find the mutation temperature of a protein. This is important since the temperature affects protein function. The spin-coupling constant provides high-level structure information about bond angles and rotation in a protein. We have used Bayesian statistics (MCMC) to find the unknown parameters for these two models. Predictive models using the parameters found with this method show good results.