Reliable prediction of protein thermostability change upon double mutation from amino acid sequence

Abstract

The accurate prediction of protein stability change upon mutation is one of the important issues for protein design. In this work, we have focused on the stability change of double mutations and systematically analyzed the wild-type and mutant residues, patterns in amino acid sequence and locations of mutants. Based on the sequence information of wild-type, mutant and three neighboring residues, we have presented a weighted decision table method (WET) for predicting the stability changes of 180 double mutants obtained from thermal (DeltaDeltaG) denaturation. Using 10-fold cross-validation test, our method showed a correlation of 0.75 between experimental and predicted values of stability changes, and an accuracy of 82.2% for discriminating the stabilizing and destabilizing mutants.

PROTS-RF: A Robust Model for Predicting Mutation-Induced Protein Stability Changes

“…These five groups are then used in a standard five-fold cross validation (CV). The second dataset includes 180 double point mutations (D180) from 27 wild-type proteins with ΔΔ G values, was collected by Huang and Gromiha [23]. The final dataset contains 141 multiple point mutations (D141) from 19 different wild type proteins which were collected from ProTherm database [34].…”

“…The model was built and tested on a set of 180 double point mutations. The correlation coefficient of the predicted and experimental ΔΔ G reached 0.75 and the overall accuracy was 82.2% in the 10-fold cross validation test [23]. However, the accuracy drops to 0.57 when it is tested on the hypothetical reverse mutations (see details in the results).…”

“…The general procedure of machine learning approaches is to train predictive models based on available experimental data using features (properties) such as substitution types, secondary structures, solvent accessibility, and the amino acid composition of neighboring residues. Many algorithms including support vector machines [17]–[20], neuronal networks [21], and multiple regression and classification techniques [22], [23], have been used for predicting protein stability changes induced by mutations. The machine learning approaches hold great promises because they may be used to discover subtle patterns governing mutation induced stability changes and protein stability in general.…”

“…Only in recent years a few studies have been focused on multiple-mutation induced thermostability changes. For example, Huang and Gromiha proposed a predictive model named WET, a weighted decision table method for predicting protein thermostability change upon double mutation from amino acid sequences [23]. The model was built and tested on a set of 180 double point mutations.…”

Real value prediction of protein folding rate change upon point mutation

“…In fact the substitution of amino acid residues in a protein alters the folding, stability, specificity and functions of proteins [29][30][31][32][33][34]. The influence of amino acid substitutions on protein stability has been studied extensively and several methods have been proposed to predict protein stability change upon mutation [24,[35][36][37][38][39][40][41][42][43][44]. On the other hand, prediction of U values based on the change in the logarithm of folding rate by the change in stability due to mutation has been reported [45,46].…”