Predicting the Effect of Single Mutations on Protein Stability and Binding with Respect to Types of Mutations

International Journal of Molecular Sciences, Journal Year: 2023, Volume and Issue: 24(15), P. 12073 - 12073

Published: July 28, 2023

The development of methods and algorithms to predict the effect mutations on protein stability, protein-protein interaction, protein-DNA/RNA binding is necessitated by needs engineering for understanding molecular mechanism disease-causing variants. vast majority leading require a database experimentally measured folding free energy changes training. These databases are collections experimental data taken from scientific investigations typically aimed at probing role particular residues above-mentioned thermodynamic characteristics, i.e., not introduced random do necessarily represent originating single nucleotide variants (SNV). Thus, reported performance assessed these or other limited cases may be applicable predicting SNVs seen in human population. Indeed, we demonstrate that non-SNVs equally presented corresponding databases, distribution same. It shown Pearson correlation coefficients (PCCs) obtained involving smaller than non-SNVs, indicating caution should used applying them reveal SNVs. Furthermore, it demonstrated some sensitive chemical nature mutations, resulting PCCs differ factor four across chemically different mutations. All found underestimate roughly 2.

Language: Английский

---

MuToN Quantifies Binding Affinity Changes upon Protein Mutations by Geometric Deep Learning

Advanced Science, Journal Year: 2024, Volume and Issue: unknown

Published: July 12, 2024

Abstract Assessing changes in protein–protein binding affinity due to mutations helps understanding a wide range of crucial biological processes within cells. Despite significant efforts create accurate computational models, predicting how affect remains challenging the complexity mechanisms involved. In present work, geometric deep learning framework called MuToN is introduced for quantifying protein change upon residue mutations. The method, designed with attention networks, mechanism‐aware. It captures interfaces mutated complexes and assesses allosteric effects amino acids. Experimental results highlight MuToN's superiority compared existing methods. Additionally, flexibility effectiveness are illustrated by its precise predictions between SARS‐CoV‐2 variants ACE2 complex.

Language: Английский

---

Graph masked self-distillation learning for prediction of mutation impact on protein–protein interactions

Communications Biology, Journal Year: 2024, Volume and Issue: 7(1)

Published: Oct. 26, 2024

Assessing mutation impact on the binding affinity change (ΔΔG) of protein-protein interactions (PPIs) plays a crucial role in unraveling structural-functional intricacies proteins and developing innovative protein designs. In this study, we present deep learning framework, PIANO, for improved prediction ΔΔG PPIs. The PIANO framework leverages graph masked self-distillation scheme structural geometric representation pre-training, which effectively captures context representations surrounding sites, makes predictions using multi-branch network consisting multiple encoders amino acids, atoms, sequences. Extensive experiments demonstrated its superior performance capability pre-trained encoder capturing meaningful representations. Compared to previous methods, can be widely applied both holo complex structures apo monomer structures. Moreover, illustrated practical applicability highlighting pathogenic mutations proteins, distinguishing de novo disease cases controls PPI systems. Overall, offers powerful tool, may provide valuable insights into study drug design, therapeutic intervention, engineering.

Language: Английский

---

Toward reporting standards for the pathogenicity of variant combinations involved in multilocus/oligogenic diseases

Human Genetics and Genomics Advances, Journal Year: 2022, Volume and Issue: 4(1), P. 100165 - 100165

Published: Dec. 2, 2022

Although standards and guidelines for the interpretation of variants identified in genes that cause Mendelian disorders have been developed, this is not case more complex genetic models including variant combinations multiple genes. During a large curation process conducted on 318 research articles presenting oligogenic combinations, we encountered several recurring issues concerning their proper reporting pathogenicity assessment. These mainly concern absence strong evidence refutes monogenic model lack functional assessment joint effect involved variants. With increasing accumulation such cases, it has become essential to develop how these oligogenic/multilocus should be interpreted, validated, reported order provide high-quality data supporting scientific community.

Language: Английский

---

Strategies to Identify Genetic Variants Causing Infertility

Trends in Molecular Medicine, Journal Year: 2021, Volume and Issue: 27(8), P. 792 - 806

Published: Jan. 12, 2021

Language: Английский

---