Decoding the functional impact of the cancer genome through protein–protein interactions

Nature reviews. Cancer, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 14, 2025

Language: Английский

---

Machine learning meets physics: A two-way street

Proceedings of the National Academy of Sciences, Journal Year: 2024, Volume and Issue: 121(27)

Published: June 24, 2024

Emotions coordinate our behavior and physiological states during survival-salient events pleasurable interactions. Even though we are often consciously aware of current emotional state, such as anger or happiness, the mechanisms giving ...Emotions felt in body, somatosensory feedback has been proposed to trigger conscious experiences. Here reveal maps bodily sensations associated with different emotions using a unique topographical self-report method. In ...

Language: Английский

---

Leveraging protein structural information to improve variant effect prediction

Current Opinion in Structural Biology, Journal Year: 2025, Volume and Issue: 92, P. 103023 - 103023

Published: Feb. 22, 2025

Despite massive sequencing efforts, understanding the difference between human pathogenic and benign variants remains a challenge. Computational variant effect predictors (VEPs) have emerged as essential tools for assessing impact of genetic variants, although their performance varies. Initially, sequence-based methods dominated field, but recent advances, particularly in protein structure prediction technologies like AlphaFold, led to an increased utilization structural information by VEPs aimed at scoring missense variants. This review highlights progress integrating into VEPs, showcasing novel models such AlphaMissense, PrimateAI-3D, CPT-1 that demonstrate improved evaluation. Structural data offers more interpretability, especially non-loss-of-function provides insights complex interactions vivo. As field utilizing biomolecular structures will be pivotal future VEP development, with breakthroughs protein-ligand protein-nucleic acid offering new avenues.

Language: Английский

---

The Historical Evolution and Significance of Multiple Sequence Alignment in Molecular Structure and Function Prediction

Biomolecules, Journal Year: 2024, Volume and Issue: 14(12), P. 1531 - 1531

Published: Nov. 29, 2024

Multiple sequence alignment (MSA) has evolved into a fundamental tool in the biological sciences, playing pivotal role predicting molecular structures and functions. With broad applications protein nucleic acid modeling, MSAs continue to underpin advancements across range of disciplines. are not only foundational for traditional comparison techniques but also increasingly important context artificial intelligence (AI)-driven advancements. Recent breakthroughs AI, particularly structure prediction, rely heavily on accuracy efficiency enhance remote homology detection guide spatial restraints. This review traces historical evolution MSA, highlighting its significance function prediction. We cover methodologies used monomers, complexes, RNA, while exploring emerging AI-based alternatives, such as language models, complementary or replacement approaches application tasks. By discussing strengths, limitations, these methods, this aims provide researchers with valuable insights MSA's evolving role, equipping them make informed decisions structural prediction research.

Language: Английский

---

DiffPaSS - High-performance differentiable pairing of protein sequences using soft scores

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 23, 2024

Abstract Identifying interacting partners from two sets of protein sequences has important applications in computational biology. Interacting share similarities across species due to their common evolutionary history, and feature correlations amino acid usage the need maintain complementary interaction interfaces. Thus, problem finding pairs can be formulated as searching for a pairing that maximizes sequence similarity or coevolution score. Several methods have been developed address this problem, applying different approximate optimization scores. We introduce DiffPaSS, framework flexible, fast, hyperparameter-free biological sequences, which applied wide variety apply it benchmark prokaryotic dataset, using mutual information neighbor graph alignment DiffPaSS outperforms existing algorithms optimizing same demonstrate usefulness our paired alignments prediction complex structure. does not require aligned, we also non-aligned T cell receptors.

Language: Английский

---

Ranking protein-peptide binding affinities with protein language models

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 15, 2024

Abstract In this study we explore the use of protein language models for ranking protein-peptide interaction strength, extending concept binary classification. We introduce a method that measures and ranks binding affinities in an unsupervised manner, eliminating need extensive labeled data, structural information, or complex biochemical features. demonstrate utility our approach across five distinct datasets by comparing predicted strength rankings with experimentally derived inhibitory concentration (IC50) values. Furthermore, discuss limitations encountered during present preliminary findings on to more general protein-protein interactions. Finally, highlight comprehensive specifically designed

Language: Английский

---

DiffPaSS—High-performance differentiable pairing of protein sequences using soft scores

Bioinformatics, Journal Year: 2024, Volume and Issue: 41(1)

Published: Dec. 13, 2024

Abstract Motivation Identifying interacting partners from two sets of protein sequences has important applications in computational biology. Interacting share similarities across species due to their common evolutionary history, and feature correlations amino acid usage the need maintain complementary interaction interfaces. Thus, problem finding pairs can be formulated as searching for a pairing that maximizes sequence similarity or coevolution score. Several methods have been developed address this problem, applying different approximate optimization scores. Results We introduce Differentiable Pairing using Soft Scores (DiffPaSS), differentiable framework flexible, fast, hyperparameter-free biological sequences, which applied wide variety apply it benchmark prokaryotic dataset, mutual information neighbor graph alignment DiffPaSS outperforms existing algorithms optimizing same demonstrate usefulness our paired alignments prediction complex structure. does not require aligned, we also nonaligned T-cell receptors. Availability implementation A PyTorch installable Python package are available at https://github.com/Bitbol-Lab/DiffPaSS.

Language: Английский

---

蛋白质复合物链间残基距离深度学习预测方法

Scientia Sinica Informationis, Journal Year: 2024, Volume and Issue: 55(1), P. 94 - 94

Published: Dec. 20, 2024

---