Critical assessment of methods of protein structure prediction (CASP)—Round XV

Proteins Structure Function and Bioinformatics, Journal Year: 2023, Volume and Issue: 91(12), P. 1539 - 1549

Published: Nov. 2, 2023

Abstract Computing protein structure from amino acid sequence information has been a long‐standing grand challenge. Critical assessment of prediction (CASP) conducts community experiments aimed at advancing solutions to this and related problems. Experiments are conducted every 2 years. The 2020 experiment (CASP14) saw major progress, with the second generation deep learning methods delivering accuracy comparable for many single proteins. There is an expectation that these will have much wider application in computational structural biology. Here we summarize results most recent experiment, CASP15, 2022, emphasis on new learning‐driven progress. Other papers special issue proteins provide more detailed analysis. For structures, AlphaFold2 method still superior other approaches, but there two points note. First, although was core all successful methods, wide variety implementation combination methods. Second, using standard protocol default parameters only produces highest quality result about thirds targets, extensive sampling required others. advance CASP enormous increase computed complexes, achieved by use overall do not fully match performance too, based perform best, again than defaults often required. Also note encouraging early compute ensembles macromolecular structures. Critically usability both derived estimates local global high quality, however interface regions slightly less reliable. CASP15 also included computation RNA structures first time. Here, classical approaches produced better agreement ones, limited. Also, time, protein–ligand area interest drug design. were ones. Many discussed conference, it clear continue advance.

Language: Английский

---

Protein target highlights in CASP15: Analysis of models by structure providers

Proteins Structure Function and Bioinformatics, Journal Year: 2023, Volume and Issue: 91(12), P. 1571 - 1599

Published: July 26, 2023

We present an in-depth analysis of selected CASP15 targets, focusing on their biological and functional significance. The authors the structures identify discuss key protein features evaluate how effectively these aspects were captured in submitted predictions. While overall ability to predict three-dimensional continues impress, reproducing uncommon not previously observed experimental is still a challenge. Furthermore, instances with conformational flexibility large multimeric complexes highlight need for novel scoring strategies better emphasize biologically relevant structural regions. Looking ahead, closer integration computational techniques will play role determining next challenges be unraveled field molecular biology.

Language: Английский

---

An end-to-end framework for the prediction of protein structure and fitness from single sequence

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Aug. 27, 2024

Significant research progress has been made in the field of protein structure and fitness prediction. Particularly, single-sequence-based prediction methods like ESMFold OmegaFold achieve a balance between inference speed accuracy, showing promise for many downstream tasks. Here, we propose SPIRED, model that exhibits comparable performance to state-of-the-art but with approximately 5-fold acceleration at least one order magnitude reduction training consumption. By integrating SPIRED neural networks, compose an end-to-end framework named SPIRED-Fitness rapid both from single sequence satisfactory accuracy. Moreover, SPIRED-Stab, derivative SPIRED-Fitness, achieves predicting mutational effects on stability. The changes caused by mutations are high interest engineering. authors develop allow high-throughput them amino acid sequence.

Language: Английский

---

Pairing interacting protein sequences using masked language modeling

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

Published: June 24, 2024

Predicting which proteins interact together from amino acid sequences is an important task. We develop a method to pair interacting protein leverages the power of language models trained on multiple sequence alignments (MSAs), such as MSA Transformer and EvoFormer module AlphaFold. formulate problem pairing partners among paralogs two families in differentiable way. introduce called Differentiable Pairing using Alignment-based Language Models (DiffPALM) that solves it by exploiting ability fill masked acids surrounding context. encodes coevolution between functionally or structurally coupled within chains. It also captures inter-chain coevolution, despite being single-chain data. Relying without fine-tuning, DiffPALM outperforms existing coevolution-based methods difficult benchmarks shallow extracted ubiquitous prokaryotic datasets. alternative based state-of-the-art model single sequences. Paired are crucial ingredient supervised deep learning predict three-dimensional structure complexes. Starting paired substantially improves prediction some eukaryotic complexes AlphaFold-Multimer. achieves competitive performance with orthology-based pairing.

Language: Английский

---

Breaking the conformational ensemble barrier: Ensemble structure modeling challenges in CASP15

Proteins Structure Function and Bioinformatics, Journal Year: 2023, Volume and Issue: 91(12), P. 1903 - 1911

Published: Oct. 23, 2023

Abstract For the first time, 2022 CASP (Critical Assessment of Structure Prediction) community experiment included a section on computing multiple conformations for protein and RNA structures. There was full or partial success in reproducing ensembles four nine targets, an encouraging result. structures, enhanced sampling with variations AlphaFold2 deep learning method by far most effective approach. One substantial conformational change caused single mutation across complex interface accurately reproduced. In two other assembly modeling cases, methods succeeded near to experimental ones even though environmental factors were not calculations. An experimentally derived flexibility ensemble allowed accurate structure model be identified. Difficulties how handle sparse low‐resolution data current lack RNA/protein complexes. However, these obstacles appear addressable.

Language: Английский

---

Physical-aware model accuracy estimation for protein complex using deep learning method

Computational and Structural Biotechnology Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

---

GDFold2: A fast and parallelizable protein folding environment with freely defined objective functions

Protein Science, Journal Year: 2025, Volume and Issue: 34(2)

Published: Jan. 28, 2025

Abstract An important step of mainstream protein structure prediction is to model the 3D based on predicted 2D inter‐residue geometric information. This folding has been integrated into a unified neural network allow end‐to‐end training in state‐of‐the‐art methods like AlphaFold2, but separately implemented using Rosetta environment some traditional trRosetta. Despite inferiority accuracy, conventional approach allows for sampling various conformations compatible with constraints, partially capturing dynamic Here, we propose GDFold2, novel environment, address limitations Rosetta. On one hand, GDFold2 highly computationally efficient, capable accomplishing multiple processes parallel within time scale minutes generic proteins. other supports freely defined objective functions fulfill diversified optimization requirements. Moreover, quality assessment (QA) provide reliable structures folded by thus substantially simplifying selection structural models. and QA could be combined investigate transition path between conformational states, online server available at https://structpred.life.tsinghua.edu.cn/server_gdfold2.html .

Language: Английский

---

Structure and function of the EDEM:PDI ERAD checkpoint complex

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 30, 2025

The ERAD glycoprotein misfolding checkpoint complex de-mannosylates misfolded glycoproteins to enable retrotranslocation, ubiquitination, and proteasomal degradation. comprises an Endoplasmic Reticulum-Degradation Enhancing α-Mannosidase (EDEM) a Protein Disulfide Isomerase (PDI). We solved Cryo-EM structures of Chaetomium thermophilum ( Ct ) CtEDEM:CtPDI, both as the heterodimer with no client in α1-antitrypsin (A1AT-NHK). EDEM catalytic domain nests within PDI arc, while A1AT-NHK binds EDEM's C-terminal flexible domains. Mass spectrometry reveals disulfide bond between exposed Cys PAD EDEM. Co-transfection EDEM, A1AT-NHK, shifts EDEM:PDI higher molecular weight non-reducing SDS-PAGE. Redox chemistry bonds generates oxidized, demannosylation-competent reduced PDI, priming function reductase, facilitating retrotranslocation.

Language: Английский

---

Initiation of ERAD by the bifunctional complex of Mnl1/Htm1 mannosidase and protein disulfide isomerase

Nature Structural & Molecular Biology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 10, 2025

Language: Английский

---

Rēs ipSAE loquunt: What′s wrong with AlphaFold′s ipTM score and how to fix it

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 14, 2025

AlphaFold's ipTM metric is used to predict the accuracy of structural predictions protein-protein interactions (PPIs) and probability that two proteins interact. Many AF2/AF3 users have experienced phenomenon if they trim full-length sequence constructs (e.g. from UniProt) interacting domains (or domain+peptide), their scores go up, even though structure prediction interaction unchanged. The reason this happens due mathematical formulation in AF2/AF3, which whole chains. If both chains a PPI complex contain large amounts disorder or accessory do not form primary domain-domain domain/peptide interaction, score can be lowered significantly. then does accurately represent nor whether actually We solved problem by: 1) including only residue pairs good predicted aligned error ( PAE ) scores; 2) by adjusting d 0 parameter (a function length query sequences) TM equation include number residues with interchain s residue; 3) using value itself distributions over calculate pairwise residue-residue pTM values into calculation. first are crucial calculating high for domain-peptide presence many hundreds disordered regions and/or domains. third allows us require common output json files AF2 AF3 (including server output) without having change AlphaFold code affecting accuracy. show benchmark new score, called ipSAE (interaction Score Aligned Errors), able separate true false complexes more efficiently than AlphaFold2's score. resulting program freely available at https://github.com/dunbracklab/IPSAE .

Language: Английский

---