Unmasking AlphaFold to integrate experiments and predictions in multimeric complexes.


Journal

Nature communications
ISSN: 2041-1723
Titre abrégé: Nat Commun
Pays: England
ID NLM: 101528555

Informations de publication

Date de publication:
09 Oct 2024
Historique:
received: 13 05 2024
accepted: 26 09 2024
medline: 9 10 2024
pubmed: 9 10 2024
entrez: 8 10 2024
Statut: epublish

Résumé

Since the release of AlphaFold, researchers have actively refined its predictions and attempted to integrate it into existing pipelines for determining protein structures. These efforts have introduced a number of functionalities and optimisations at the latest Critical Assessment of protein Structure Prediction edition (CASP15), resulting in a marked improvement in the prediction of multimeric protein structures. However, AlphaFold's capability of predicting large protein complexes is still limited and integrating experimental data in the prediction pipeline is not straightforward. In this study, we introduce AF_unmasked to overcome these limitations. Our results demonstrate that AF_unmasked can integrate experimental information to build larger or hard to predict protein assemblies with high confidence. The resulting predictions can help interpret and augment experimental data. This approach generates high quality (DockQ score > 0.8) structures even when little to no evolutionary information is available and imperfect experimental structures are used as a starting point. AF_unmasked is developed and optimised to fill incomplete experimental structures (structural inpainting), which may provide insights into protein dynamics. In summary, AF_unmasked provides an easy-to-use method that efficiently integrates experiments to predict large protein complexes more confidently.

Identifiants

pubmed: 39379372
doi: 10.1038/s41467-024-52951-w
pii: 10.1038/s41467-024-52951-w
doi:

Substances chimiques

Proteins 0

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

8724

Subventions

Organisme : Science for Life Laboratory (SciLifeLab)
ID : BeyondFold
Organisme : Knut och Alice Wallenbergs Stiftelse (Knut and Alice Wallenberg Foundation)
ID : KAW2021.0347

Informations de copyright

© 2024. The Author(s).

Références

Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
pubmed: 34265844 pmcid: 8371605 doi: 10.1038/s41586-021-03819-2
Roney, J. P. & Ovchinnikov, S. State-of-the-art estimation of protein model accuracy using AlphaFold. Phys. Rev. Lett. 129, 238101 (2022).
pubmed: 36563190 doi: 10.1103/PhysRevLett.129.238101
Evans, R. et al. Protein complex prediction with AlphaFold-Multimer. BioRxiv https://doi.org/10.1101/2021.10.04.463034 (2021).
Wallner, B. AFsample: Improving multimer prediction with alphafold using massive sampling. Bioinformatics 39, btad573 (2023).
Committee, C. CASP15: Book of Abstracts. https://predictioncenter.org/casp15/doc/CASP15_Abstracts.pdf (2022).
Bryant, P. et al. Predicting the structure of large protein complexes using Alphafold and Monte Carlo tree search. Nat. Commun. 13, 6028 (2022).
pubmed: 36224222 pmcid: 9556563 doi: 10.1038/s41467-022-33729-4
Perrakis, A. & Sixma, T. K. AI revolutions in biology: the joys and perils of AlphaFold. EMBO Rep. 22, e54046 (2021).
pubmed: 34668287 pmcid: 8567224 doi: 10.15252/embr.202154046
Terwilliger, T. C. et al. AlphaFold predictions are valuable hypotheses, and accelerate but do not replace experimental structure determination. bioRxiv https://doi.org/10.1101/2022.11.21.517405 (2022).
Liebschner, D. Using predicted models in Phenix. Acta Cryst. 75, 861–877 (2019).
Terwilliger, T. C. et al. Accelerating crystal structure determination with iterative AlphaFold prediction. Acta Crystallogr. Sec. D. Struct. Biol. 79, 234–242 (2023).
doi: 10.1107/S205979832300102X
Terwilliger, T. C. et al. Improved AlphaFold modeling with implicit experimental information. Nat. Methods 19, 1376–1382 (2022).
pubmed: 36266465 pmcid: 9636017 doi: 10.1038/s41592-022-01645-6
Ahdritz, G. et al. OpenFold: Retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization. bioRxiv https://www.biorxiv.org/content/10.1101/2022.11.20.517210 (2022).
Li, Z. et al. Uni-fold: An open-source platform for developing protein folding models beyond Alphafold. bioRxiv https://www.biorxiv.org/content/10.1101/2022.08.04.502811v3.full.pdf (2022).
Stahl, K., Brock, O. & Rappsilber, J. Modelling protein complexes with crosslinking mass spectrometry and deep learning. bioRxiv https://doi.org/10.1101/2023.06.07.544059 (2023).
Mirabello, C. & Wallner, B. DockQv2: Improved automatic quality measure forprotein multimers, nucleic acids and small molecules. Bioinformatics. btae586 https://doi.org/10.1093/bioinformatics/btae586 (2024).
Yin, R. & Pierce, B. G. Evaluation of Alphafold antibody-antigen modeling with implications for improving predictive accuracy. bioRxiv 33, e4865 (2023).
Gray, J. J. et al. Protein–protein docking with simultaneous optimization of rigid-body displacement and side-chain conformations. J. Mol. Biol. 331, 281–299 (2003).
pubmed: 12875852 doi: 10.1016/S0022-2836(03)00670-3
Lensink, M. F. & Wodak, S. J. Docking, scoring, and affinity prediction in Capri. Proteins Struct. Funct. Bioinforma. 81, 2082–2095 (2013).
doi: 10.1002/prot.24428
Basu, S. & Wallner, B. DockQ: a quality measure for protein-protein docking models. PLoS ONE 11, e0161879 (2016).
pubmed: 27560519 pmcid: 4999177 doi: 10.1371/journal.pone.0161879
Zidek, A. AlphaFold v2.3.0 Release Notes and CASP15 Models. https://github.com/deepmind/alphafold/blob/main/docs/technical_note_v2.3.0.md (2022).
Pak, M. A. et al. Using AlphaFold to predict the impact of single mutations on protein stability and function. PLoS ONE 18, e0282689 (2023).
pubmed: 36928239 pmcid: 10019719 doi: 10.1371/journal.pone.0282689
Mao, Y. et al. The small subunit of Rubisco and its potential as an engineering target. J. Exp. Bot. 74, 543–561 (2023).
pubmed: 35849331 doi: 10.1093/jxb/erac309
Valegård, K., Hasse, D., Andersson, I. & Gunn, L. H. Structure of rubisco from arabidopsis thaliana in complex with 2-carboxyarabinitol-1,5-bisphosphate. Acta Crystallogr. D. Struct. Biol. 74, 1–9 (2018).
pubmed: 29372894 pmcid: 5786004 doi: 10.1107/S2059798317017132
Yin, Y. et al. Structural basis for aggregate dissolution and refolding by the mycobacterium tuberculosis ClpB-DnaK bi-chaperone system. Cell Rep. 35, 109166 (2021).
pubmed: 34038719 pmcid: 8209680 doi: 10.1016/j.celrep.2021.109166
Kędzierska-Mieszkowska, S. & Zolkiewski, M. Hsp100 molecular chaperone clpb and its role in virulence of bacterial pathogens. Int. J. Mol. Sci. 22, 5319 (2021).
pubmed: 34070174 pmcid: 8158500 doi: 10.3390/ijms22105319
Katikaridis, P., Bohl, V. & Mogk, A. Resisting the heat: bacterial disaggregases rescue cells from devastating protein aggregation. Front. Mol. Biosci. 8, 681439 (2021).
pubmed: 34017857 pmcid: 8129007 doi: 10.3389/fmolb.2021.681439
Avellaneda, M. J. et al. Processive extrusion of polypeptide loops by a hsp100 disaggregase. Nature 578, 317–320 (2020).
pubmed: 31996849 doi: 10.1038/s41586-020-1964-y
Mazal, H., Iljina, M., Riven, I. & Haran, G. Ultrafast pore-loop dynamics in a AAA+ machine point to a Brownian-ratchet mechanism for protein translocation. Sci. Adv. 7, eabg4674 (2021).
pubmed: 34516899 pmcid: 8442866 doi: 10.1126/sciadv.abg4674
Uchihashi, T. et al. Dynamic structural states of ClpB involved in its disaggregation function. Nat. Commun. 9, 2147 (2018).
pubmed: 29858573 pmcid: 5984625 doi: 10.1038/s41467-018-04587-w
Deville, C., Franke, K., Mogk, A., Bukau, B. & Saibil, H. R. Two-step activation mechanism of the ClpB disaggregase for sequential substrate threading by the main ATPase motor. Cell Rep. 27, 3433–3446.e4 (2019).
pubmed: 31216466 pmcid: 6593972 doi: 10.1016/j.celrep.2019.05.075
Rizo, A. N. et al. Structural basis for substrate gripping and translocation by the ClpB AAA+ disaggregase. Nat. Commun. 10, 1–12 (2019).
doi: 10.1038/s41467-019-10150-y
Yu, H. et al. ATP hydrolysis-coupled peptide translocation mechanism of Mycobacterium tuberculosis ClpB. Proc. Natl Acad. Sci. USA 115, E9560–E9569 (2018).
pubmed: 30257943 pmcid: 6187150 doi: 10.1073/pnas.1810648115
Deville, C. et al. Structural pathway of regulated substrate transfer and threading through an hsp100 disaggregase. Sci. Adv. 3, e1701726 (2017).
pubmed: 28798962 pmcid: 5544394 doi: 10.1126/sciadv.1701726
Haslberger, T. et al. M domains couple the ClpB threading motor with the DnaK chaperone activity. Mol. Cell 25, 247–260 (2007).
pubmed: 17244532 doi: 10.1016/j.molcel.2006.11.008
Oguchi, Y. et al. A tightly regulated molecular toggle controls AAA+ disaggregase. Nat. Struct. Mol. Biol. 19, 1338–1346 (2012).
pubmed: 23160353 doi: 10.1038/nsmb.2441
Rosenzweig, R., Moradi, S., Zarrine-Afsar, A., Glover, J. R. & Kay, L. E. Unraveling the mechanism of protein disaggregation through a ClpB-DnaK interaction. Science 339, 1080–1083 (2013).
pubmed: 23393091 doi: 10.1126/science.1233066
Carroni, M. et al. Head-to-tail interactions of the coiled-coil domains regulate ClpB activity and cooperation with hsp70 in protein disaggregation. Elife 3, e02481 (2014).
pubmed: 24843029 pmcid: 4023160 doi: 10.7554/eLife.02481
Mazal, H. et al. Tunable microsecond dynamics of an allosteric switch regulate the activity of a AAA+ disaggregation machine. Nat. Commun. 10, 1438 (2019).
pubmed: 30926805 pmcid: 6440998 doi: 10.1038/s41467-019-09474-6
Rosenzweig, R. et al. Clpb n-terminal domain plays a regulatory role in protein disaggregation. Proc. Natl Acad. Sci. USA 112, E6872–E6881 (2015).
pubmed: 26621746 pmcid: 4687599 doi: 10.1073/pnas.1512783112
Bergoug, M. et al. Neurofibromin structure, functions and regulation. Cells 9, 2365 (2020).
pubmed: 33121128 pmcid: 7692384 doi: 10.3390/cells9112365
Gutmann, D. H., Wood, D. L. & Collins, F. S. Identification of the neurofibromatosis type 1 gene product. Proc. Natl Acad. Sci. USA 88, 9658–9662 (1991).
pubmed: 1946382 pmcid: 52777 doi: 10.1073/pnas.88.21.9658
Ratner, N. & Miller, S. J. A RASopathy gene commonly mutated in cancer: the neurofibromatosis type 1 tumour suppressor. Nat. Rev. Cancer 15, 290–301 (2015).
pubmed: 25877329 pmcid: 4822336 doi: 10.1038/nrc3911
Naschberger, A., Baradaran, R., Rupp, B. & Carroni, M. The structure of neurofibromin isoform 2 reveals different functional states. Nature 599, 315–319 (2021).
pubmed: 34707296 pmcid: 8580823 doi: 10.1038/s41586-021-04024-x
Lupton, C. J. et al. The cryo-EM structure of the human neurofibromin dimer reveals the molecular basis for neurofibromatosis type 1. Nat. Struct. Mol. Biol. 28, 982–988 (2021).
pubmed: 34887559 doi: 10.1038/s41594-021-00687-2
Chaker-Margot, M. et al. Structural basis of activation of the tumor suppressor protein neurofibromin. Mol. Cell 82, 1288–1296.e5 (2022).
Young, L. C. et al. Destabilizing NF1 variants act in a dominant negative manner through neurofibromin dimerization. Proc. Natl Acad. Sci. USA 120, e2208960120 (2023).
pubmed: 36689660 pmcid: 9945959 doi: 10.1073/pnas.2208960120
Zhang, Y. & Skolnick, J. TM-align: a protein structure alignment algorithm based on the tm-score. Nucleic acids Res. 33, 2302–2309 (2005).
pubmed: 15849316 pmcid: 1084323 doi: 10.1093/nar/gki524
Mirabello, C. lDDT-align: A Tool to Align Protein Structures While Maximizing lddt. https://github.com/clami66/lDDT_align (2022).
Li, Z., Nguyen, S. P., Xu, D. & Shang, Y. Protein loop modeling using deep generative adversarial network. In 2017 IEEE 29th International Conference on Tools with Artificial Intelligence (ICTAI). 1085–1091 (IEEE, 2017).
Anand, N. & Huang, P. Generative modeling for protein structures. Adv. Neural Inform. Process. Syst. https://doi.org/10.1038/s41587-023-02115-w (2018).
Wang, J. et al. Scaffolding protein functional sites using deep learning. Science 377, 387–394 (2022).
pubmed: 35862514 pmcid: 9621694 doi: 10.1126/science.abn2100
Baek, M. et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science 373, 871–876 (2021).
pubmed: 34282049 pmcid: 7612213 doi: 10.1126/science.abj8754
Lai, B., McPartlon, M. & Xu, J. End-to-end deep structure generative model for protein design. bioRxiv https://doi.org/10.1101/2022.07.09.499440 (2022).
Zhang, C. et al. Framedipt: Se (3) diffusion model for protein structure inpainting. bioRxiv https://doi.org/10.1101/2023.11.21.568057 (2023).
Collins, K. W. et al. Dockground resource for protein recognition studies. Protein Sci. 31, e4481 (2022).
pubmed: 36281025 pmcid: 9667896 doi: 10.1002/pro.4481
Mukherjee, S. & Zhang, Y. Mm-align: a quick algorithm for aligning multiple-chain protein complex structures using iterative dynamic programming. Nucleic Acids Res. 37, e83–e83 (2009).
pubmed: 19443443 pmcid: 2699532 doi: 10.1093/nar/gkp318
van Kempen, M. et al. Foldseek: fast and accurate protein structure search. Biorxiv https://doi.org/10.1101/2022.02.07.479398 (2022).
Wallner, B. Improved multimer prediction using massive sampling with Alphafold in casp15. Proteins Struct. Function Bioinform. 91, 1734–1746 (2023).
Zhang, Y. & Skolnick, J. Scoring function for automated assessment of protein structure template quality. Proteins Struct. Funct. Bioinform. 57, 702–710 (2004).
doi: 10.1002/prot.20264
Pintilie, G. et al. Measurement of atom resolvability in cryo-em maps with q-scores. Nat. Methods 17, 328–334 (2020).
pubmed: 32042190 pmcid: 7446556 doi: 10.1038/s41592-020-0731-1
Mirabello, C. Alphafold unmasked: integration of experiments and predictions in multimeric complexes. Zenodo https://doi.org/10.5281/zenodo.13364959 (2024).

Auteurs

Claudio Mirabello (C)

Dept of Physics, Chemistry and Biology, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Linköping University, 581 83, Linköping, Sweden. claudio.mirabello@scilifelab.se.

Björn Wallner (B)

Dept of Physics, Chemistry and Biology, Linköping University, 581 83, Linköping, Sweden.

Björn Nystedt (B)

Dept of Cell and Molecular Biology, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Uppsala University, Husargatan 3, SE-752 37, Uppsala, Sweden.

Stavros Azinas (S)

Dept of Biochemistry and Biophysics, Science for Life Laboratory, Stockholm University, Stockholm, Sweden.

Marta Carroni (M)

Dept of Biochemistry and Biophysics, Science for Life Laboratory, Stockholm University, Stockholm, Sweden.

Articles similaires

Selecting optimal software code descriptors-The case of Java.

Yegor Bugayenko, Zamira Kholmatova, Artem Kruglov et al.
1.00
Software Algorithms Programming Languages
Databases, Protein Protein Domains Protein Folding Proteins Deep Learning

Exploring blood-brain barrier passage using atomic weighted vector and machine learning.

Yoan Martínez-López, Paulina Phoobane, Yanaima Jauriga et al.
1.00
Blood-Brain Barrier Machine Learning Humans Support Vector Machine Software
1.00
Humans Magnetic Resonance Imaging Brain Infant, Newborn Infant, Premature

Classifications MeSH