Bioinformatics tools that predict protein stability changes upon point mutations have made a lot of progress in the last decades and have become accurate and fast enough to make computational mutagenesis experiments feasible, even on a proteome scale. Despite these achievements, they still suffer from important issues that must be solved to allow further improving their performances and utilizing them to deepen our insights into protein folding and stability mechanisms. One of these problems is their bias toward the learning datasets which, being dominated by destabilizing mutations, causes predictions to be better for destabilizing than for stabilizing mutations. We thoroughly analyzed the biases in the prediction of folding free energy changes upon point mutations (⁠$ΔΔG0$⁠) and proposed some unbiased solutions. We started by constructing a dataset $Ssym$ of experimentally measured $ΔΔG0$s with an equal number of stabilizing and destabilizing mutations, by collecting mutations for which the structure of both the wild-type and mutant protein is available. On this balanced dataset, we assessed the performances of 15 widely used $ΔΔG0$ predictors. After the astonishing observation that almost all these methods are strongly biased toward destabilizing mutations, especially those that use black-box machine learning, we proposed an elegant way to solve the bias issue by imposing physical symmetries under inverse mutations on the model structure, which we implemented in PoPMuSiC^sym. This new predictor constitutes an efficient trade-off between accuracy and absence of biases. Some final considerations and suggestions for further improvement of the predictors are discussed. Supplementary data are available at Bioinformatics online. The article 10.1093/bioinformatics/bty340/, published alongside this paper, also addresses the problem of biases in protein stability change predictions.

The accurate prediction of the impact of an amino acid substitution on the thermal stability of a protein is a central issue in protein science, and is of key relevance for the rational optimization of various bioprocesses that use enzymes in unusual conditions. Here we present one of the first computational tools to predict the change in melting temperature ΔT_m upon point mutations, given the protein structure and, when available, the melting temperature T_m of the wild-type protein. The key ingredients of our model structure are standard and temperature-dependent statistical potentials, which are combined with the help of an artificial neural network. The model structure was chosen on the basis of a detailed thermodynamic analysis of the system. The parameters of the model were identified on a set of more than 1,600 mutations with experimentally measured ΔT_m. The performance of our method was tested using a strict 5-fold cross-validation procedure, and was found to be significantly superior to that of competing methods. We obtained a root mean square deviation between predicted and experimental ΔT_m values of 4.2 °C that reduces to 2.9 °C when ten percent outliers are removed. A webserver-based tool is freely available for non-commercial use at soft.dezyme.com.

Accurate in silico prediction of conformational B-cell epitopes would lead to major improvements in disease diagnostics, drug design and vaccine development. A variety of computational methods, mainly based on machine learning approaches, have been developed in the last decades to tackle this challenging problem. Here, we rigorously benchmarked nine state-of-the-art conformational B-cell epitope prediction webservers, including generic and antibody-specific methods, on a dataset of over 250 antibody-antigen structures. The results of our assessment and statistical analyses show that all the methods achieve very low performances, and some do not perform better than randomly generated patches of surface residues. In addition, we also found that commonly used consensus strategies that combine the results from multiple webservers are at best only marginally better than random. Finally, we applied all the predictors to the SARS-CoV-2 spike protein as an independent case study, and showed that they perform poorly in general, which largely recapitulates our benchmarking conclusions. We hope that these results will lead to greater caution when using these tools until the biases and issues that limit current methods have been addressed, promote the use of state-of-the-art evaluation methodologies in future publications and suggest new strategies to improve the performance of conformational B-cell epitope prediction methods.

The molecular bases of protein stability remain far from elucidated even though substantial progress has been made through both computational and experimental investigations. One of the most challenging goals is the development of accurate prediction tools of the temperature dependence of the standard folding free energy ΔG(T). Such predictors have an enormous series of potential applications, which range from drug design in the biopharmaceutical sector to the optimization of enzyme activity for biofuel production. There is thus an important demand for novel, reliable and fast predictors. We present the SCooP algorithm, which is a significant step towards accurate temperature-dependent stability prediction. This automated tool uses the protein structure and the host organism as sole entries and predicts the full T-dependent stability curve of monomeric proteins assumed to follow a two-state folding transition. Equivalently, it predicts all the thermodynamic quantities associated to the folding transition, namely the melting temperature T_m, the standard folding enthalpy ΔH_m measured at T_m, and the standard folding heat capacity ΔC_p. The cross-validated performances are good, with correlation coefficients between predicted and experimental values equal to [0.80, 0.83, 0.72] for ΔH_m, ΔC_p and T_m, respectively, which increase up to [0.88, 0.90, 0.78] upon the removal of 10% outliers. Moreover, the stability curve prediction of a target protein is very fast: it takes less than a minute. SCooP can thus potentially be applied on a structurome scale. This opens new perspectives of large-scale analyses of protein stability, which is of considerable interest for protein engineering. The SCooP webserver is freely available at http://babylone.ulb.ac.be/SCooP. Supplementary data are available at Bioinformatics online.

The classification of human genetic variants into deleterious and neutral is a challenging issue, whose complexity is rooted in the large variety of biophysical mechanisms that can be responsible for disease conditions. For non-synonymous mutations in structured proteins, one of these is the protein stability change, which can lead to loss of protein structure or function. We developed a stability-driven knowledge-based classifier that uses protein structure, artificial neural networks and solvent accessibility-dependent combinations of statistical potentials to predict whether destabilizing or stabilizing mutations are disease-causing. Our predictor yields a balanced accuracy of 71% in cross validation. As expected, it has a very high positive predictive value of 89%: it predicts with high accuracy the subset of mutations that are deleterious because of stability issues, but is by construction unable of classifying variants that are deleterious for other reasons. Its combination with an evolutionary-based predictor increases the balanced accuracy up to 75%, and allowed predicting more than 1/4 of the variants with 95% positive predictive value. Our method, called SNPMuSiC, can be used with both experimental and modeled structures and compares favorably with other prediction tools on several independent test sets. It constitutes a step towards interpreting variant effects at the molecular scale. SNPMuSiC is freely available at https://soft.dezyme.com/.