Before and after AlphaFold2: An overview of protein structure prediction

Frontiers in Bioinformatics, Journal Year: 2023, Volume and Issue: 3

Published: Feb. 28, 2023

Three-dimensional protein structure is directly correlated with its function and determination critical to understanding biological processes addressing human health life science problems in general. Although new structures are experimentally obtained over time, there still a large difference between the number of sequences placed Uniprot those resolved tertiary structure. In this context, studies have emerged predict by methods based on template or free modeling. last years, different been combined overcome their individual limitations, until emergence AlphaFold2, which demonstrated that predicting high accuracy at unprecedented scale possible. Despite current impact field, AlphaFold2 has limitations. Recently, language models promised revolutionize structural biology allowing discovery only from evolutionary patterns present sequence. Even though these do not reach accuracy, they already covered some being able more than 200 million proteins metagenomic databases. mini-review, we provide an overview breakthroughs prediction before after emergence.

Language: Английский

---

AFsample: improving multimer prediction with AlphaFold using massive sampling

Bioinformatics, Journal Year: 2023, Volume and Issue: 39(9)

Published: Sept. 1, 2023

Abstract Summary The AlphaFold2 neural network model has revolutionized structural biology with unprecedented performance. We demonstrate that by stochastically perturbing the enabling dropout at inference combined massive sampling, it is possible to improve quality of generated models. ∼6000 models per target compared 25 default for AlphaFold-Multimer, v1 and v2 multimer models, without templates, increased number recycles within network. method was benchmarked in CASP15, AlphaFold-Multimer improved average DockQ from 0.41 0.55 using identical input ranked very top protein assembly category when all other groups participating CASP15. simplicity should facilitate adaptation field, be useful anyone interested modeling multimeric structures, alternate conformations, or flexible structures. Availability implementation AFsample available online http://wallnerlab.org/AFsample.

Language: Английский

---

Progress at protein structure prediction, as seen in CASP15

Current Opinion in Structural Biology, Journal Year: 2023, Volume and Issue: 80, P. 102594 - 102594

Published: April 14, 2023

In Dec 2020, the results of AlphaFold version 2 were presented at CASP14, sparking a revolution in field protein structure predictions. For first time, purely computational method could challenge experimental accuracy for prediction single domains. The code v2 was released summer 2021, and since then, it has been shown that can be used to accurately predict most ordered proteins many protein–protein interactions. It also sparked an explosion development field, improving AI-based methods complexes, disordered regions, design. Here I will review some inventions by release AlphaFold.

Language: Английский

---

AlphaFold, allosteric, and orthosteric drug discovery: Ways forward

Drug Discovery Today, Journal Year: 2023, Volume and Issue: 28(6), P. 103551 - 103551

Published: March 11, 2023

Drug discovery is arguably a highly challenging and significant interdisciplinary aim. The stunning success of the artificial intelligence-powered AlphaFold, whose latest version buttressed by an innovative machine-learning approach that integrates physical biological knowledge about protein structures, raised drug hopes unsurprisingly, have not come to bear. Even though accurate, models are rigid, including pockets. AlphaFold's mixed performance poses question how its power can be harnessed in discovery. Here we discuss possible ways going forward wielding strengths, while bearing mind what AlphaFold cannot do. For kinases receptors, input enriched active (ON) state better chance rational design success.

Language: Английский

---

PEP-FOLD4: a pH-dependent force field for peptide structure prediction in aqueous solution

Nucleic Acids Research, Journal Year: 2023, Volume and Issue: 51(W1), P. W432 - W437

Published: May 11, 2023

Abstract Accurate and fast structure prediction of peptides less 40 amino acids in aqueous solution has many biological applications, but their conformations are pH- salt concentration-dependent. In this work, we present PEP-FOLD4 which goes one step beyond machine-learning approaches, such as AlphaFold2, TrRosetta RaptorX. Adding the Debye-Hueckel formalism for charged-charged side chain interactions to a Mie all intramolecular (backbone chain) interactions, PEP-FOLD4, based on coarse-grained representation peptides, performs well methods well-structured displays significant improvements poly-charged peptides. is available at http://bioserv.rpbs.univ-paris-diderot.fr/services/PEP-FOLD4. This server free there no login requirement.

Language: Английский

---

Evaluation of AlphaFold antibody–antigen modeling with implications for improving predictive accuracy

Protein Science, Journal Year: 2023, Volume and Issue: 33(1)

Published: Dec. 11, 2023

Abstract High resolution antibody–antigen structures provide critical insights into immune recognition and can inform therapeutic design. The challenges of experimental structural determination the diversity repertoire underscore necessity accurate computational tools for modeling complexes. Initial benchmarking showed that despite overall success in protein–protein complexes, AlphaFold AlphaFold‐Multimer have limited interactions. In this study, we performed a thorough analysis AlphaFold's performance on 427 nonredundant complex structures, identifying useful confidence metrics predicting model quality, features complexes associated with improved success. Notably, found latest version improves near‐native to over 30%, versus approximately 20% previous version, while increased sampling gives 50% With success, generate models many cases, additional training or other optimization may further improve performance.

Language: Английский

---

From interaction networks to interfaces, scanning intrinsically disordered regions using AlphaFold2

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

Published: Jan. 18, 2024

The revolution brought about by AlphaFold2 opens promising perspectives to unravel the complexity of protein-protein interaction networks. analysis networks obtained from proteomics experiments does not systematically provide delimitations regions. This is particular concern in case interactions mediated intrinsically disordered regions, which site generally small. Using a dataset protein-peptide complexes involving regions that are non-redundant with structures used training, we show when using full sequences proteins, AlphaFold2-Multimer only achieves 40% success rate identifying correct and structure interface. By delineating region into fragments decreasing size combining different strategies for integrating evolutionary information, manage raise this up 90%. We obtain similar rates much larger protein taken ELM database. Beyond identification site, our study also explores specificity issues. advantages limitations confidence score discriminate between alternative binding partners, task can be particularly challenging small motifs.

Language: Английский

---

CombFold: predicting structures of large protein assemblies using a combinatorial assembly algorithm and AlphaFold2

Nature Methods, Journal Year: 2024, Volume and Issue: 21(3), P. 477 - 487

Published: Feb. 7, 2024

Abstract Deep learning models, such as AlphaFold2 and RosettaFold, enable high-accuracy protein structure prediction. However, large complexes are still challenging to predict due their size the complexity of interactions between multiple subunits. Here we present CombFold, a combinatorial hierarchical assembly algorithm for predicting structures utilizing pairwise subunits predicted by AlphaFold2. CombFold accurately (TM-score >0.7) 72% among top-10 predictions in two datasets 60 large, asymmetric assemblies. Moreover, structural coverage was 20% higher compared corresponding Protein Data Bank entries. We applied method on from Complex Portal with known stoichiometry but without obtained high-confidence predictions. supports integration distance restraints based crosslinking mass spectrometry fast enumeration possible complex stoichiometries. CombFold’s high accuracy makes it promising tool expanding beyond monomeric proteins.

Language: Английский

---

Structure prediction of alternative protein conformations

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

Published: Aug. 26, 2024

Abstract Proteins are dynamic molecules whose movements result in different conformations with functions. Neural networks such as AlphaFold2 can predict the structure of single-chain proteins most likely to exist PDB. However, almost all protein structures multiple represented PDB have been used while training these models. Therefore, it is unclear whether alternative be genuinely predicted using networks, or if they simply reproduced from memory. Here, we train a prediction network, Cfold, on conformational split generate conformations. Cfold enables efficient exploration landscape monomeric structures. Over 50% experimentally known nonredundant evaluated here high accuracy (TM-score > 0.8).

Language: Английский

---

Enhanced Protein-Protein Interaction Discovery via AlphaFold-Multimer

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

Published: Feb. 21, 2024

Abstract Accurately mapping protein-protein interactions (PPIs) is critical for elucidating cellular functions and has significant implications health disease. Conventional experimental approaches, while foundational, often fall short in capturing direct, dynamic interactions, especially those with transient or small interfaces. Our study leverages AlphaFold-Multimer (AFM) to re-evaluate high-confidence PPI datasets from Drosophila human. analysis uncovers a limitation of the AFM-derived interface pTM (ipTM) metric, which, reflective structural integrity, can miss physiologically relevant at interfaces within flexible regions. To bridge this gap, we introduce Local Interaction Score (LIS), derived AFM’s Predicted Aligned Error (PAE), focusing on areas low PAE values, indicative high confidence interaction predictions. The LIS method demonstrates enhanced sensitivity detecting PPIs, particularly among that involve By applying large-scale datasets, enhance detection direct interactions. Moreover, present FlyPredictome, an online platform integrates our AFM-based predictions additional information such as gene expression correlations subcellular localization This not only improves upon utility prediction but also highlights potential computational methods complement approaches identification networks.

Language: Английский

---