Assessing computational methods for predicting protein stability upon mutation: good on average but not in the details.

By Vladimir Potapov, Mati Cohen, Gideon Schreiber. 2009

346 citations recorded by the source corpus.

Subfields: Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

In Molecular Biology

Citations in the corpus

Role of conformational sampling in computing mutation-induced changes in protein structure and stability (2010, methodology): “Prompted by a recent study reporting poor performance of the Rosetta methodology in predicting the free energy changes associated with mutations, 7 we present here a detailed analysis of the tradeoff between the resolution of the energy function and the extent of conformational sampling in ΔΔ G…”
Variation Interpretation Predictors: Principles, Types, Performance, and Choice (2016, methodology): “Six programs were compared [Potapov et al., 2009] by using 2,156 single variations from ProTherm (Table 2).”
Mutational effects on stability are largely conserved during protein evolution (2013, background): “Even state-ofthe-art force fields are at best modestly accurate at predicting mutational effects (6, 7).”
Functional Constraint Profiling of a Viral Protein Reveals Discordance of Evolutionary Conservation and Functionality (2015, background): “It has been shown that although most force fields exhibit a correct trend in ΔΔG prediction, the precision is still lacking as compared to experimental methods [73].”
Tools for Predicting the Functional Impact of Nonsynonymous Genetic Variation (2016, methodology): “Using this criterion, Potapov et al. found that the accuracy of predicting such hot spots was between 72 and 80% across six different commonly used methods (Potapov et al. 2009).”
PoPMuSiC 2.1: a web server for the estimation of protein stability changes upon mutation and sequence optimality (2011, neutral): “Three recent studies independently assessed and compared the performances of several of those predictors, using datasets of experimentally characterized mutants that had not been used to train any of the predictive models [22,29,30].”
A Web Resource for Standardized Benchmark Datasets, Metrics, and Rosetta Protocols for Macromolecular Modeling and Design (2015, background): “[16]; a set of 1210 mutants collected by Kellogg et al.”
How Structural and Physicochemical Determinants Shape Sequence Constraints in a Functional Enzyme (2015, methodology): “These are examples of constraints that act to stabilize superficial hydrophobic clusters, as opposed to those clusters truly at the protein core which are better described by dependencies on amino acid volume, steric hindrance and FoldX predictions (see next subsection).”
Blind tests of RNA–protein binding affinity prediction (2019, background): “The best meth-ods for these other macromolecular interactions report accuracies of between 1 and 2 kcal/mol and include rigorous blind studies, validating their use for applications that range from drug discovery to protein – protein interface design (14, 16, 17).”
Maximum-Entropy Models of Sequenced Immune Repertoires Predict Antigen-Antibody Affinity (2016, neutral): “Currently, available methods are time demanding and, more importantly, their predictions are hard to assess [2, 3].”
Epistasis in protein evolution (2016, background): “Computational simulations suggest that both smalland large-effect epistasis have a pervasive influence on proteins’ evolutionary trajectories(6); however, the models used in these simulations are highly simplified, so the real-world relevance of their quantitative conclusions is unclear.(67) Experim”