1
|
Durairaj J, Waterhouse AM, Mets T, Brodiazhenko T, Abdullah M, Studer G, Tauriello G, Akdel M, Andreeva A, Bateman A, Tenson T, Hauryliuk V, Schwede T, Pereira J. Uncovering new families and folds in the natural protein universe. Nature 2023; 622:646-653. [PMID: 37704037 PMCID: PMC10584680 DOI: 10.1038/s41586-023-06622-3] [Citation(s) in RCA: 13] [Impact Index Per Article: 13.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/24/2023] [Accepted: 09/07/2023] [Indexed: 09/15/2023]
Abstract
We are now entering a new era in protein sequence and structure annotation, with hundreds of millions of predicted protein structures made available through the AlphaFold database1. These models cover nearly all proteins that are known, including those challenging to annotate for function or putative biological role using standard homology-based approaches. In this study, we examine the extent to which the AlphaFold database has structurally illuminated this 'dark matter' of the natural protein universe at high predicted accuracy. We further describe the protein diversity that these models cover as an annotated interactive sequence similarity network, accessible at https://uniprot3d.org/atlas/AFDB90v4 . By searching for novelties from sequence, structure and semantic perspectives, we uncovered the β-flower fold, added several protein families to Pfam database2 and experimentally demonstrated that one of these belongs to a new superfamily of translation-targeting toxin-antitoxin systems, TumE-TumA. This work underscores the value of large-scale efforts in identifying, annotating and prioritizing new protein families. By leveraging the recent deep learning revolution in protein bioinformatics, we can now shed light into uncharted areas of the protein universe at an unprecedented scale, paving the way to innovations in life sciences and biotechnology.
Collapse
Affiliation(s)
- Janani Durairaj
- Biozentrum, University of Basel, Basel, Switzerland
- SIB Swiss Institute of Bioinformatics, University of Basel, Basel, Switzerland
| | - Andrew M Waterhouse
- Biozentrum, University of Basel, Basel, Switzerland
- SIB Swiss Institute of Bioinformatics, University of Basel, Basel, Switzerland
| | - Toomas Mets
- Institute of Technology, University of Tartu, Tartu, Estonia
- Department of Experimental Medical Science, Lund University, Lund, Sweden
| | | | - Minhal Abdullah
- Institute of Technology, University of Tartu, Tartu, Estonia
- Department of Experimental Medical Science, Lund University, Lund, Sweden
| | - Gabriel Studer
- Biozentrum, University of Basel, Basel, Switzerland
- SIB Swiss Institute of Bioinformatics, University of Basel, Basel, Switzerland
| | - Gerardo Tauriello
- Biozentrum, University of Basel, Basel, Switzerland
- SIB Swiss Institute of Bioinformatics, University of Basel, Basel, Switzerland
| | | | - Antonina Andreeva
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, UK
| | - Alex Bateman
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, UK
| | - Tanel Tenson
- Institute of Technology, University of Tartu, Tartu, Estonia
| | - Vasili Hauryliuk
- Institute of Technology, University of Tartu, Tartu, Estonia
- Department of Experimental Medical Science, Lund University, Lund, Sweden
- Science for Life Laboratory, Lund, Sweden
- Virus Centre, Lund University, Lund, Sweden
| | - Torsten Schwede
- Biozentrum, University of Basel, Basel, Switzerland.
- SIB Swiss Institute of Bioinformatics, University of Basel, Basel, Switzerland.
| | - Joana Pereira
- Biozentrum, University of Basel, Basel, Switzerland.
- SIB Swiss Institute of Bioinformatics, University of Basel, Basel, Switzerland.
| |
Collapse
|
2
|
Ferruz N, Heinzinger M, Akdel M, Goncearenco A, Naef L, Dallago C. From sequence to function through structure: Deep learning for protein design. Comput Struct Biotechnol J 2022; 21:238-250. [PMID: 36544476 PMCID: PMC9755234 DOI: 10.1016/j.csbj.2022.11.014] [Citation(s) in RCA: 18] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/31/2022] [Revised: 11/05/2022] [Accepted: 11/05/2022] [Indexed: 11/20/2022] Open
Abstract
The process of designing biomolecules, in particular proteins, is witnessing a rapid change in available tooling and approaches, moving from design through physicochemical force fields, to producing plausible, complex sequences fast via end-to-end differentiable statistical models. To achieve conditional and controllable protein design, researchers at the interface of artificial intelligence and biology leverage advances in natural language processing (NLP) and computer vision techniques, coupled with advances in computing hardware to learn patterns from growing biological databases, curated annotations thereof, or both. Once learned, these patterns can be used to provide novel insights into mechanistic biology and the design of biomolecules. However, navigating and understanding the practical applications for the many recent protein design tools is complex. To facilitate this, we 1) document recent advances in deep learning (DL) assisted protein design from the last three years, 2) present a practical pipeline that allows to go from de novo-generated sequences to their predicted properties and web-powered visualization within minutes, and 3) leverage it to suggest a generated protein sequence which might be used to engineer a biosynthetic gene cluster to produce a molecular glue-like compound. Lastly, we discuss challenges and highlight opportunities for the protein design field.
Collapse
Key Words
- ADMM, Alternating Direction Method of Multipliers
- CNN, Convolutional Neural Network
- DL, Deep learning
- Deep learning
- Drug discovery
- FNN, fully-connected neural network
- GAN, Generative Adversarial Network
- GCN, Graph Convolutional Network
- GNN, Graph Neural Network
- GO, Gene Ontology
- GVP, Geometric Vector Perceptron
- LSTM, Long-Short Term Memory
- MLP, Multilayer Perceptron
- MSA, Multiple Sequence Alignment
- NLP, Natural Language Processing
- NSR, Natural Sequence Recovery
- Protein design
- Protein language models
- Protein prediction
- VAE, Variational Autoencoder
- pLM, protein Language Model
Collapse
Affiliation(s)
- Noelia Ferruz
- Institute of Informatics and Applications, University of Girona, Girona, Spain
- Department of Biochemistry, University of Bayreuth, Bayreuth, Germany
| | - Michael Heinzinger
- Department of Informatics, Bioinformatics & Computational Biology, Technische Universität München, 85748 Garching, Germany
| | - Mehmet Akdel
- VantAI, 151 W 42nd Street, New York, NY 10036, United States
| | | | - Luca Naef
- VantAI, 151 W 42nd Street, New York, NY 10036, United States
| | - Christian Dallago
- Department of Informatics, Bioinformatics & Computational Biology, Technische Universität München, 85748 Garching, Germany
- VantAI, 151 W 42nd Street, New York, NY 10036, United States
- NVIDIA DE GmbH, Einsteinstraße 172, 81677 München, Germany
| |
Collapse
|
3
|
Akdel M, Pires DEV, Pardo EP, Jänes J, Zalevsky AO, Mészáros B, Bryant P, Good LL, Laskowski RA, Pozzati G, Shenoy A, Zhu W, Kundrotas P, Serra VR, Rodrigues CHM, Dunham AS, Burke D, Borkakoti N, Velankar S, Frost A, Basquin J, Lindorff-Larsen K, Bateman A, Kajava AV, Valencia A, Ovchinnikov S, Durairaj J, Ascher DB, Thornton JM, Davey NE, Stein A, Elofsson A, Croll TI, Beltrao P. A structural biology community assessment of AlphaFold2 applications. Nat Struct Mol Biol 2022; 29:1056-1067. [PMID: 36344848 PMCID: PMC9663297 DOI: 10.1038/s41594-022-00849-w] [Citation(s) in RCA: 176] [Impact Index Per Article: 88.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/25/2021] [Accepted: 09/20/2022] [Indexed: 11/09/2022]
Abstract
Most proteins fold into 3D structures that determine how they function and orchestrate the biological processes of the cell. Recent developments in computational methods for protein structure predictions have reached the accuracy of experimentally determined models. Although this has been independently verified, the implementation of these methods across structural-biology applications remains to be tested. Here, we evaluate the use of AlphaFold2 (AF2) predictions in the study of characteristic structural elements; the impact of missense variants; function and ligand binding site predictions; modeling of interactions; and modeling of experimental structural data. For 11 proteomes, an average of 25% additional residues can be confidently modeled when compared with homology modeling, identifying structural features rarely seen in the Protein Data Bank. AF2-based predictions of protein disorder and complexes surpass dedicated tools, and AF2 models can be used across diverse applications equally well compared with experimentally determined structures, when the confidence metrics are critically considered. In summary, we find that these advances are likely to have a transformative impact in structural biology and broader life-science research.
Collapse
Affiliation(s)
- Mehmet Akdel
- Bioinformatics Group, Department of Plant Sciences, Wageningen University and Research, Wageningen, the Netherlands
| | - Douglas E V Pires
- School of Computing and Information Systems, University of Melbourne, Melbourne, Victoria, Australia
| | - Eduard Porta Pardo
- Josep Carreras Leukaemia Research Institute (IJC), Badalona, Spain
- Barcelona Supercomputing Center (BSC), Barcelona, Spain
| | - Jürgen Jänes
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - Arthur O Zalevsky
- Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russian Federation
| | | | - Patrick Bryant
- Dep of Biochemistry and Biophysics and Science for Life Laboratory, Solna, Sweden
| | - Lydia L Good
- Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark
| | - Roman A Laskowski
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - Gabriele Pozzati
- Dep of Biochemistry and Biophysics and Science for Life Laboratory, Solna, Sweden
| | - Aditi Shenoy
- Dep of Biochemistry and Biophysics and Science for Life Laboratory, Solna, Sweden
| | - Wensi Zhu
- Dep of Biochemistry and Biophysics and Science for Life Laboratory, Solna, Sweden
| | - Petras Kundrotas
- Dep of Biochemistry and Biophysics and Science for Life Laboratory, Solna, Sweden
| | | | - Carlos H M Rodrigues
- School of Computing and Information Systems, University of Melbourne, Melbourne, Victoria, Australia
| | - Alistair S Dunham
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - David Burke
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - Neera Borkakoti
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - Sameer Velankar
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - Adam Frost
- Department of Biochemistry and Biophysics University of California, San Francisco, CA, USA
| | - Jérôme Basquin
- Department of Structural Cell Biology, Max Planck Institute of Biochemistry, Martinsried, Germany
| | - Kresten Lindorff-Larsen
- Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark
| | - Alex Bateman
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK
| | - Andrey V Kajava
- Université de Montpellier, Centre de Recherche en Biologie Cellulaire de Montpellier (CRBM) CNRS, Montpellier, France
| | | | - Sergey Ovchinnikov
- Faculty of Arts and Sciences, Division of Science, Harvard University, Cambridge, MA, USA.
| | | | - David B Ascher
- School of Chemistry and Molecular Biology, University of Queensland, Brisbane, Queensland, Australia.
| | - Janet M Thornton
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK.
| | | | - Amelie Stein
- Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark.
| | - Arne Elofsson
- Dep of Biochemistry and Biophysics and Science for Life Laboratory, Solna, Sweden.
| | - Tristan I Croll
- Cambridge Institute for Medical Research, Department of Haematology, The University of Cambridge, Cambridge, UK.
| | - Pedro Beltrao
- European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Cambridge, UK.
- Institute of Molecular Systems Biology, ETH Zürich, Zürich, Switzerland.
| |
Collapse
|
4
|
Durairaj J, Akdel M, de Ridder D, van Dijk ADJ. Geometricus represents protein structures as shape-mers derived from moment invariants. Bioinformatics 2021; 36:i718-i725. [PMID: 33381814 DOI: 10.1093/bioinformatics/btaa839] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 09/15/2020] [Indexed: 01/28/2023] Open
Abstract
MOTIVATION As the number of experimentally solved protein structures rises, it becomes increasingly appealing to use structural information for predictive tasks involving proteins. Due to the large variation in protein sizes, folds and topologies, an attractive approach is to embed protein structures into fixed-length vectors, which can be used in machine learning algorithms aimed at predicting and understanding functional and physical properties. Many existing embedding approaches are alignment based, which is both time-consuming and ineffective for distantly related proteins. On the other hand, library- or model-based approaches depend on a small library of fragments or require the use of a trained model, both of which may not generalize well. RESULTS We present Geometricus, a novel and universally applicable approach to embedding proteins in a fixed-dimensional space. The approach is fast, accurate, and interpretable. Geometricus uses a set of 3D moment invariants to discretize fragments of protein structures into shape-mers, which are then counted to describe the full structure as a vector of counts. We demonstrate the applicability of this approach in various tasks, ranging from fast structure similarity search, unsupervised clustering and structure classification across proteins from different superfamilies as well as within the same family. AVAILABILITY AND IMPLEMENTATION Python code available at https://git.wur.nl/durai001/geometricus.
Collapse
Affiliation(s)
| | - Mehmet Akdel
- Bioinformatics Group, Department of Plant Sciences
| | | | - Aalt D J van Dijk
- Bioinformatics Group, Department of Plant Sciences.,Mathematical and Statistical Methods - Biometris, Department of Plant Sciences, Wageningen University and Research, Wageningen 6700AP, The Netherlands
| |
Collapse
|
5
|
Akdel M, Durairaj J, de Ridder D, van Dijk ADJ. Caretta - A multiple protein structure alignment and feature extraction suite. Comput Struct Biotechnol J 2020; 18:981-992. [PMID: 32368333 PMCID: PMC7186369 DOI: 10.1016/j.csbj.2020.03.011] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/14/2019] [Revised: 02/01/2020] [Accepted: 03/13/2020] [Indexed: 02/06/2023] Open
Abstract
The vast number of protein structures currently available opens exciting opportunities for machine learning on proteins, aimed at predicting and understanding functional properties. In particular, in combination with homology modelling, it is now possible to not only use sequence features as input for machine learning, but also structure features. However, in order to do so, robust multiple structure alignments are imperative. Here we present Caretta, a multiple structure alignment suite meant for homologous but sequentially divergent protein families which consistently returns accurate alignments with a higher coverage than current state-of-the-art tools. Caretta is available as a GUI and command-line application and additionally outputs an aligned structure feature matrix for a given set of input structures, which can readily be used in downstream steps for supervised or unsupervised machine learning. We show Caretta’s performance on two benchmark datasets, and present an example application of Caretta in predicting the conformational state of cyclin-dependent kinases.
Collapse
Affiliation(s)
- Mehmet Akdel
- Bioinformatics Group, Department of Plant Sciences, Wageningen University and Research, The Netherlands
| | - Janani Durairaj
- Bioinformatics Group, Department of Plant Sciences, Wageningen University and Research, The Netherlands
| | - Dick de Ridder
- Bioinformatics Group, Department of Plant Sciences, Wageningen University and Research, The Netherlands
| | - Aalt D J van Dijk
- Bioinformatics Group, Department of Plant Sciences, Wageningen University and Research, The Netherlands.,Mathematical and Statistical Methods - Biometris, Department of Plant Sciences, Wageningen University and Research, The Netherlands
| |
Collapse
|
6
|
Abstract
MOTIVATION Next-generation sequencing technology is generating a wealth of highly similar genome sequences for many species, paving the way for a transition from single-genome to pan-genome analyses. Accordingly, genomics research is going to switch from reference-centric to pan-genomic approaches. We define the pan-genome as a comprehensive representation of multiple annotated genomes, facilitating analyses on the similarity and divergence of the constituent genomes at the nucleotide, gene and genome structure level. Current pan-genomic approaches do not thoroughly address scalability, functionality and usability. RESULTS We introduce a generalized De Bruijn graph as a pan-genome representation, as well as an online algorithm to construct it. This representation is stored in a Neo4j graph database, which makes our approach scalable to large eukaryotic genomes. Besides the construction algorithm, our software package, called PanTools, currently provides functionality for annotating pan-genomes, adding sequences, grouping genes, retrieving gene sequences or genomic regions, reconstructing genomes and comparing and querying pan-genomes. We demonstrate the performance of the tool using datasets of 62 E. coli genomes, 93 yeast genomes and 19 Arabidopsis thaliana genomes. AVAILABILITY AND IMPLEMENTATION The Java implementation of PanTools is publicly available at http://www.bif.wur.nl CONTACT sandra.smit@wur.nl.
Collapse
Affiliation(s)
- Siavash Sheikhizadeh
- Bioinformatics Group, Wageningen University, Droevendaalsesteeg 1, 6708PB, Wageningen, The Netherlands
| | - M Eric Schranz
- Biosystematics Group, Wageningen University, Droevendaalsesteeg 1, 6708PB, The Netherlands
| | - Mehmet Akdel
- Bioinformatics Group, Wageningen University, Droevendaalsesteeg 1, 6708PB, Wageningen, The Netherlands
| | - Dick de Ridder
- Bioinformatics Group, Wageningen University, Droevendaalsesteeg 1, 6708PB, Wageningen, The Netherlands
| | - Sandra Smit
- Bioinformatics Group, Wageningen University, Droevendaalsesteeg 1, 6708PB, Wageningen, The Netherlands
| |
Collapse
|