Predicting protein stability changes upon single-point mutation: a thorough comparison of the available tools on a new dataset.

Predicting the difference in thermodynamic stability between protein variants is crucial for protein design and understanding the genotype-phenotype relationships. So far, several computational tools have been created to address this task. Nevertheless, most of them have been trained or optimized on the same and 'all' available data, making a fair comparison unfeasible. Here, we introduce a novel dataset, collected and manually cleaned from the latest version of the ThermoMutDB database, consisting of 669 variants not included in the most widely used training datasets. The prediction performance and the ability to satisfy the antisymmetry property by considering both direct and reverse variants were evaluated across 21 different tools. The Pearson correlations of the tested tools were in the ranges of 0.21-0.5 and 0-0.45 for the direct and reverse variants, respectively. When both direct and reverse variants are considered, the antisymmetric methods perform better achieving a Pearson correlation in the range of 0.51-0.62. The tested methods seem relatively insensitive to the physiological conditions, performing well also on the variants measured with more extreme pH and temperature values. A common issue with all the tested methods is the compression of the $\Delta \Delta G$ predictions toward zero. Furthermore, the thermodynamic stability of the most significantly stabilizing variants was found to be more challenging to predict. This study is the most extensive comparisons of prediction methods using an entirely novel set of variants never tested before.

Assessing the performance of computational predictors for estimating protein stability changes upon missense mutations.

Understanding how a mutation might affect protein stability is of significant importance to protein engineering and for understanding protein evolution genetic diseases. While a number of computational tools have been developed to predict the effect of missense mutations on protein stability protein stability upon mutations, they are known to exhibit large biases imparted in part by the data used to train and evaluate them. Here, we provide a comprehensive overview of predictive tools, which has provided an evolving insight into the importance and relevance of features that can discern the effects of mutations on protein stability. A diverse selection of these freely available tools was benchmarked using a large mutation-level blind dataset of 1342 experimentally characterised mutations across 130 proteins from ThermoMutDB, a second test dataset encompassing 630 experimentally characterised mutations across 39 proteins from iStable2.0 and a third blind test dataset consisting of 268 mutations in 27 proteins from the newly published ProThermDB. The performance of the methods was further evaluated with respect to the site of mutation, type of mutant residue and by ranging the pH and temperature. Additionally, the classification performance was also evaluated by classifying the mutations as stabilizing (∆∆G ≥ 0) or destabilizing (∆∆G < 0). The results reveal that the performance of the predictors is affected by the site of mutation and the type of mutant residue. Further, the results show very low performance for pH values 6-8 and temperature higher than 65 for all predictors except iStable2.0 on the S630 dataset. To illustrate how stability and structure change upon single point mutation, we considered four stabilizing, two destabilizing and two stabilizing mutations from two proteins, namely the toxin protein and bovine liver cytochrome. Overall, the results on S268, S630 and S1342 datasets show that the performance of the integrated predictors is better than the mechanistic or individual machine learning predictors. We expect that this paper will provide useful guidance for the design and development of next-generation bioinformatic tools for predicting protein stability changes upon mutations.

Cataract-Associated New Mutants S175G/H181Q of ßΒ2-Crystallin and P24S/S31G of γD-Crystallin Are Involved in Protein Aggregation by Structural Changes.

ß/γ-Crystallins, the main structural protein in human lenses, have highly stable structure for keeping the lens transparent. Their mutations have been linked to cataracts. In this study, we identified 10 new mutations of ß/γ-crystallins in lens proteomic dataset of cataract patients using bioinformatics tools. Of these, two double mutants, S175G/H181Q of ßΒ2-crystallin and P24S/S31G of γD-crystallin, were found mutations occurred in the largest loop linking the distant ß-sheets in the Greek key motif. We selected these double mutants for identifying the properties of these mutations, employing biochemical assay, the identification of protein modifications with nanoUPLC-ESI-TOF tandem MS and examining their structural dynamics with hydrogen/deuterium exchange-mass spectrometry (HDX-MS). We found that both double mutations decrease protein stability and induce the aggregation of ß/γ-crystallin, possibly causing cataracts. This finding suggests that both the double mutants can serve as biomarkers of cataracts.

Are machine learning based methods suited to address complex biological problems? Lessons from CAGI-5 challenges.

In silico approaches are routinely adopted to predict the effects of genetic variants and their relation to diseases. The critical assessment of genome interpretation (CAGI) has established a common framework for the assessment of available predictors of variant effects on specific problems and our group has been an active participant of CAGI since its first edition. In this paper, we summarize our experience and lessons learned from the last edition of the experiment (CAGI-5). In particular, we analyze prediction performances of our tools on five CAGI-5 selected challenges grouped into three different categories: prediction of variant effects on protein stability, prediction of variant pathogenicity, and prediction of complex functional effects. For each challenge, we analyze in detail the performance of our tools, highlighting their potentialities and drawbacks. The aim is to better define the application boundaries of each tool.

Changes in temperament over time in young children with autism spectrum disorder and other developmental delays: a follow-up comparison study.

BACKGROUND: Although in the last decade some research has emerged on temperament in autism spectrum disorder (ASD), this research has primarily focused on the differences between children with ASD and their typically developing peers rather than the stability or change in temperament in this population. Thus, the goal of this study was to examine temperament over time in children with ASD, developmental delays (DD) and typical development (TD). Temperament differences were also compared among the three groups. METHODS: To accomplish this, parents rated children's temperament at Time 1 (T1) and Time 2 (T2) using the Carey Temperament Scales (CTS). RESULTS: Results from the study showed that at T1, parents of children with ASD rated their children as more withdrawn (i.e. approach), and emotionally negative (i.e. mood), and less distractible and adaptable than parents of children with TD and DD. Also, children with ASD were rated as more intense and children with DD as less distractible than their TD peers. Similarly, at T2, children with ASD were rated more withdrawn, and emotionally negative, and less persistent, rhythmic, adaptable and distractible than children with TD and DD. Also, children with ASD were rated as more active than their DD peers. Regarding stability, parent ratings of temperament appeared stable over time in the TD group, but ratings varied substantially in the ASD or DD groups. That is, for the ASD group, activity and approach at T1 were significantly associated with their corresponding dimensions at T2. However, for the TD group, rhythmicity, approach, intensity and mood at T1 were significantly associated with those dimensions at T2. No associations were found in the DD group. Regarding change, parents reported change in rhythmicity, persistence and threshold between T1 and T2 in the ASD group. Similarly, parents reported change in rhythmicity, approach and threshold between T1 and T2 in the DD group. Lastly, parents of TD children reported change in adaptability, persistence and distractibility between T1 and T2. CONCLUSIONS: These findings are novel in that children with ASD appear to have less stable temperament profile and different change patterns than children with TD or DD. Similar to previous research, children with ASD were described by their parents as experiencing more temperamental difficulties.

ProS-GNN: Predicting effects of mutations on protein stability using graph neural networks.

Predicting protein stability change upon variation through a computational approach is a valuable tool to unveil the mechanisms of mutation-induced drug failure and develop immunotherapy strategies. Some previous machine learning-based techniques exhibit anti-symmetric bias toward destabilizing situations, whereas others struggle with generalization to unseen examples. To address these issues, we propose a gated graph neural network-based approach to predict changes in protein stability upon mutation. The model uses message passing to encode the links between the molecular structure and property after eliminating the non-mutant structure and creating input feature vectors. While doing so, it also incorporates the coordinates of the raw atoms to provide spatial insights into the chemical systems. We test the model on the Ssym, Myoglobin, Broom, and p53 datasets to demonstrate the generalization performance. Compared to existing approaches, our proposed method achieves improved linearity with symmetry in less time. The code for this study is available at: https://github.com/HongzhouTang/Pros-GNN.

Influence of Model Structures on Predictors of Protein Stability Changes from Single-Point Mutations.

Missense variation in genomes can affect protein structure stability and, in turn, the cell physiology behavior. Predicting the impact of those variations is relevant, and the best-performing computational tools exploit the protein structure information. However, most of the current protein sequence variants are unresolved, and comparative or ab initio tools can provide a structure. Here, we evaluate the impact of model structures, compared to experimental structures, on the predictors of protein stability changes upon single-point mutations, where no significant changes are expected between the original and the mutated structures. We show that there are substantial differences among the computational tools. Methods that rely on coarse-grained representation are less sensitive to the underlying protein structures. In contrast, tools that exploit more detailed molecular representations are sensible to structures generated from comparative modeling, even on single-residue substitutions.

BayeStab: Predicting effects of mutations on protein stability with uncertainty quantification.

Predicting protein thermostability change upon mutation is crucial for understanding diseases and designing therapeutics. However, accurately estimating Gibbs free energy change of the protein remained a challenge. Some methods struggle to generalize on examples with no homology and produce uncalibrated predictions. Here we leverage advances in graph neural networks for protein feature extraction to tackle this structure-property prediction task. Our method, BayeStab, is then tested on four test datasets, including S669, S611, S350, and Myoglobin, showing high generalization and symmetry performance. Meanwhile, we apply concrete dropout enabled Bayesian neural networks to infer plausible models and estimate uncertainty. By decomposing the uncertainty into parts induced by data noise and model, we demonstrate that the probabilistic method allows insights into the inherent noise of the training datasets, which is closely relevant to the upper bound of the task. Finally, the BayeStab web server is created and can be found at: http://www.bayestab.com. The code for this work is available at: https://github.com/HongzhouTang/BayeStab.

iStable 2.0: Predicting protein thermal stability changes by integrating various characteristic modules.

Protein mutations can lead to structural changes that affect protein function and result in disease occurrence. In protein engineering, drug design or and optimization industries, mutations are often used to improve protein stability or to change protein properties while maintaining stability. To provide possible candidates for novel protein design, several computational tools for predicting protein stability changes have been developed. Although many prediction tools are available, each tool employs different algorithms and features. This can produce conflicting prediction results that make it difficult for users to decide upon the correct protein design. Therefore, this study proposes an integrated prediction tool, iStable 2.0, which integrates 11 sequence-based and structure-based prediction tools by machine learning and adds protein sequence information as features. Three coding modules are designed for the system, an Online Server Module, a Stand-alone Module and a Sequence Coding Module, to improve the prediction performance of the previous version of the system. The final integrated structure-based classification model has a higher Matthews correlation coefficient than that of the single prediction tool (0.708 vs 0.547, respectively), and the Pearson correlation coefficient of the regression model likewise improves from 0.669 to 0.714. The sequence-based model not only successfully integrates off-the-shelf predictors but also improves the Matthews correlation coefficient of the best single prediction tool by at least 0.161, which is better than the individual structure-based prediction tools. In addition, both the Sequence Coding Module and the Stand-alone Module maintain performance with only a 5% decrease of the Matthews correlation coefficient when the integrated online tools are unavailable. iStable 2.0 is available at http://ncblab.nchu.edu.tw/iStable2.

Good and bad children in metabolic networks.

Equilibrium bifurcations arise from sign changes of Jacobian determinants, as parameters are varied. Therefore we address the Jacobian determinant for metabolic networks with general reaction kinetics. Our approach is based on the concept of Child Selections: each (mother) metabolite is mapped, injectively, to one of those (child) reactions that it drives as an input. Our analysis distinguishes reaction network Jacobians with constant sign from the bifurcation case, where that sign depends on specific reaction rates. In particular, we distinguish "good" Child Selections, which do not affect the sign, from more interesting and mischievous "bad" children, which gang up towards sign changes, instability, and bifurcation.

Recognition Dynamics of Cancer Mutations on the ERp57-Tapasin Interface.

Down regulation of the major histocompatibility class (MHC) I pathway plays an important role in tumour development, and can be achieved by suppression of HLA expression or mutations in the MHC peptide-binding pocket. The peptide-loading complex (PLC) loads peptides on the MHC-I molecule in a dynamic multi-step assembly process. The effects of cancer variants on ERp57 and tapasin components from the MHC-I pathway is less known, and they could have an impact on antigen presentation. Applying computational approaches, we analysed whether the ERp57-tapasin binding might be altered by missense mutations. The variants H408R(ERp57) and P96L, D100A, G183R(tapasin) at the protein-protein interface improved protein stability (ΔΔG) during the initial screen of 14 different variants. The H408R(ERp57) and P96L(tapasin) variants, located close to disulphide bonds, were further studied by molecular dynamics (MD). Identifying intramolecular a-a' domain interactions, MD revealed open and closed conformations of ERp57 in the presence and absence of tapasin. In wild-type and mutant ERp57-tapasin complexes, residues Val97, Ser98, Tyr100, Trp405, Gly407(ERp57) and Asn94, Cys95, Arg97, Asp100(tapasin) formed common H-bond interactions. Moreover, comparing the H-bond networks for P96L and H408R with each other, suggests that P96L(tapasin) improved ERp57-tapasin binding more than the H408R(ERp57) mutant. During MD, the C-terminus domain (that binds MHC-I) in tapasin from the ERp57(H408R)-tapasin complex moved away from the PLC, whereas in the ERp57-tapasin(P96L) system was oppositely displaced. These findings can have implications for the function of PLC and, ultimately, for the presentation of MHC-I peptide complex on the tumour cell surface.