The DOI matches its Crossref record; 274 of the 278 references it cites were found in a registry or an index.
Protein representations: Encoding biological information for machine learning in biocatalysis
Open accessPublished version of a preprint
Abstract
Enzymes offer a more environmentally friendly and low-impact solution to conventional chemistry, but they often require additional engineering for their application in industrial settings, an endeavour that is challenging and laborious. To address this issue, the power of machine learning can be harnessed to produce predictive models that enable the in silico study and engineering of improved enzymatic properties. Such machine learning models, however, require the conversion of the complex biological information to a numerical input, also called protein representations. These inputs demand special attention to ensure the training of accurate and precise models, and, in this review, we therefore examine the critical step of encoding protein information to numeric representations for use in machine learning. We selected the most important approaches for encoding the three distinct biological protein representations — primary sequence, 3D structure, and dynamics — to explore their requirements for employment and inductive biases. Combined representations of proteins and substrates are also introduced as emergent tools in biocatalysis. We propose the division of fixed representations, a collection of rule-based encoding strategies, and learned representations extracted from the latent spaces of large neural networks. To select the most suitable protein representation, we propose two main factors to consider. The first one is the model setup, which is influenced by the size of the training dataset and the choice of architecture. The second factor is the model objectives such as consideration about the assayed property, the difference between wild-type models and mutant predictors, and requirements for explainability. This review is aimed at serving as a source of information and guidance for properly representing enzymes in future machine learning models for biocatalysis. • Robust protein representations serve as crucial links between biology and data science for machine learning in biocatalysis. • Fixed representations impose specific inherent biases on the protein encoding through rule-based descriptors. • Learned representations from self-supervised deep learning models offer valuable biological information for supervised tasks. • Despite the ability to capture the temporal dimensions, protein dynamics representations are underexplored for enzyme models. • Choosing an encoding strategy requires considering the “model setup” and “model objective” to obtain the best representation.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Items 201–274 of 274
Using BLOSUM in Sequence AlignmentsPreprint
Mount, D. W. (2008) · Cold Spring Harbor Protocols · doi:10.1101/pdb.top39 · Cited by 63
Ensuring the Sustainability of Biocatalysis
Woodley, J. M. (2022) · ChemSusChem · doi:10.1002/cssc.202102683 · Cited by 62
Gergel, S., Soler, J., Klein, A., Schülke, K. H., Hauer, B., Garcia‐Borràs, M., & Hammer, S. C. (2023) · Nature Catalysis · doi:10.1038/s41929-023-00979-4 · Cited by 62
ProT-VAE: Protein Transformer Variational AutoEncoder for functional protein design
Sevgen, E., Moller, J., Lange, A., Parker, J., Quigley, S. D., Mayer, J., Srivastava, P., Gayatri, S., Hosfield, D. J., Dilks, C. M., et al. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2408737122 · Cited by 59
HSEpred: predict half-sphere exposure from protein sequences
Song, J., Tan, H., Takemoto, K., & Akutsu, T. (2008) · Bioinformatics · doi:10.1093/bioinformatics/btn222 · Cited by 58
Biosensor and machine learning-aided engineering of an amaryllidaceae enzyme
d’Oelsnitz, S., Diaz, D. J., Kim, W., Acosta, D. J., Dangerfield, T. L., Schechter, M. W., Minus, M. B., Howard, J. R., Do, H., Loy, J. M., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-46356-y · Cited by 58
Enzyme Dynamics: Looking Beyond a Single Structure
Agarwal, P. K., Bernard, D. N., Bafna, K., & Doucet, N. (2020) · ChemCatChem · doi:10.1002/cctc.202000665 · Cited by 57
Improved Bst DNA Polymerase Variants Derived via a Machine Learning Approach
Paik, I., Ngo, P. H. T., Shroff, R., Diaz, D. J., Maranhao, A. C., Walker, D. J. F., Bhadra, S., & Ellington, A. D. (2021) · Biochemistry · doi:10.1021/acs.biochem.1c00451 · Cited by 57
Time Evolution of the Millisecond Allosteric Activation of Imidazole Glycerol Phosphate Synthase
Calvó‐Tusell, C., Maria‐Solano, M. A., Osuna, S., & Feixas, F. (2022) · Journal of the American Chemical Society · doi:10.1021/jacs.1c12629 · Cited by 54
Wang, Y., Wei, Z., & Xi, L. (2022) · BMC Bioinformatics · doi:10.1186/s12859-022-04762-3 · Cited by 53
Ghorbani, M., Prasad, S., Klauda, J. B., & Brooks, B. R. (2022) · The Journal of Chemical Physics · doi:10.1063/5.0085607 · Cited by 52
Protein Ensemble Generation Through Variational Autoencoder Latent Space Sampling
Mansoor, S., Baek, M., Park, H., Lee, G. R., & Baker, D. A. (2024) · Journal of Chemical Theory and Computation · doi:10.1021/acs.jctc.3c01057 · Cited by 52
Robinson, S. L., Smith, M. D., Richman, J. E., Aukema, K. G., & Wackett, L. P. (2020) · Synthetic Biology · doi:10.1093/synbio/ysaa004 · Cited by 51
Tuñón, I., Laage, D., & Hynes, J. T. (2015) · Archives of Biochemistry and Biophysics · doi:10.1016/j.abb.2015.06.004 · Cited by 50
Carlin, D. A., Caster, R. W., Wang, X., Betzenderfer, S. A., Chen, C. X., Duong, V. M., Ryklansky, C. V., Alpekin, A., Beaumont, N., Kapoor, H., et al. (2016) · PLoS ONE · doi:10.1371/journal.pone.0147596 · Cited by 49
Li, M., Wang, H., Yang, Z., Zhang, L., & Zhu, Y. (2023) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2023.11.006 · Cited by 49
Computational remodeling of an enzyme conformational landscape for altered substrate selectivity
St‐Jacques, A. D., Rodriguez, J. M., Eason, M. G., Foster, S. M., Khan, S. T., Damry, A. M., Goto, N. K., Thompson, M. C., & Chica, R. A. (2023) · Nature Communications · doi:10.1038/s41467-023-41762-0 · Cited by 48
Röttig, M., Rausch, C. R., & Kohlbacher, O. (2010) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1000636 · Cited by 47
DLKcat cannot predict meaningful kcat values for mutants and unfamiliar enzymes
Kroll, A., & Lercher, M. J. (2024) · Biology Methods and Protocols · doi:10.1093/biomethods/bpae061 · Cited by 47
Audagnotto, M., Czechtizky, W., Maria, L. D., Käck, H., Papoian, G. A., Tornberg, L., Tyrchan, C., & Ulander, J. (2022) · Scientific Reports · doi:10.1038/s41598-022-13714-z · Cited by 44
Data‐Driven Protein Engineering for Improving Catalytic Activity and Selectivity
Ao, Y., Dörr, M., Menke, M. J., Born, S., Heuson, E., & Bornscheuer, U. T. (2023) · ChemBioChem · doi:10.1002/cbic.202300754 · Cited by 44
Oliveira, A. S. F., Ciccotti, G., Haider, S. M., & Mulholland, A. J. (2021) · The European Physical Journal B · doi:10.1140/epjb/s10051-021-00157-0 · Cited by 43
The shortest path method (SPM) webserver for computational enzyme design
Casadevall, G., Casadevall, J., Duran, C., & Osuna, S. (2024) · Protein Engineering Design and Selection · doi:10.1093/protein/gzae005 · Cited by 42
Therapeutic enzyme engineering using a generative neural network
Giessel, A. J., Dousis, A. D., Ravichandran, K. R., Smith, K., Sur, S., McFadyen, I. J., Zheng, W., & Licht, S. (2022) · Scientific Reports · doi:10.1038/s41598-022-05195-x · Cited by 41
Iqbal, S., Ge, F., Li, F., Akutsu, T., Zheng, Y., Gasser, R. B., Yu, D., Webb, G. I., & Song, J. (2022) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.2c00799 · Cited by 41
Castelli, M., Marchetti, F., Osuna, S., Oliveira, A. S. F., Mulholland, A. J., Serapian, S. A., & Colombo, G. (2023) · Journal of the American Chemical Society · doi:10.1021/jacs.3c11396 · Cited by 40
Li, G., Qin, Y., Fontaine, N., Chong, M. N. F., Maria‐Solano, M. A., Feixas, F., Cadet, X. F., Pandjaitan, R., Garcia‐Borràs, M., Cadet, F., et al. (2020) · ChemBioChem · doi:10.1002/cbic.202000612 · Cited by 37
EMNGly: predicting N-linked glycosylation sites using the language models for feature extraction
Hou, X., Wang, Y., Bu, D., Wang, Y., & Sun, S. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad650 · Cited by 37
Latent Biases in Machine Learning Models for Predicting Binding Affinities Using Popular Data Sets
Kanakala, G. C., Aggarwal, R., Nayar, D., & Priyakumar, U. D. (2023) · ACS Omega · doi:10.1021/acsomega.2c06781 · Cited by 36
Novel Pet‐Degrading Enzymes: Structure‐Function from a Computational Perspective
Berselli, A., Ramos, M. J., & Menziani, M. C. (2021) · ChemBioChem · doi:10.1002/cbic.202000841 · Cited by 36
Vasina, M., Vaňáček, P., Hon, J., Kovar, D. R., Faldynová, H., Kunka, A., Buryška, T., Badenhorst, C. P. S., Mazurenko, S., Bednář, D., et al. (2022) · Chem Catalysis · doi:10.1016/j.checat.2022.09.011 · Cited by 36
Self-citationMulti-Scale Representation Learning on ProteinsPreprint
Somnath, V. R., Bunne, C., & Krause, A. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2204.02337 · Cited by 34
Huang, T.-W., Zaretzki, J. M., Bergeron, C., Bennett, K. P., & Breneman, C. M. (2013) · Journal of Chemical Information and Modeling · doi:10.1021/ci4004688 · Cited by 33
Wapeesittipan, P., Mey, A. S. J. S., Walkinshaw, M. D., & Michel, J. (2019) · Communications Chemistry · doi:10.1038/s42004-019-0136-1 · Cited by 33
PON-Sol2: Prediction of Effects of Variants on Protein Solubility
Yang, Y., Zeng, L., & Vihinen, M. (2021) · International Journal of Molecular Sciences · doi:10.3390/ijms22158027 · Cited by 32
Chai, M., Moradi, S., Erfani, E., Asadnia, M., Chen, V., & Razmjou, A. (2021) · Chemistry of Materials · doi:10.1021/acs.chemmater.1c02476 · Cited by 32
Venanzi, N. A. E., Basciu, A., Vargiu, A. V., Kiparissides, A., Dalby, P. A., & Dikicioǧlu, D. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.3c00999 · Cited by 32
Target-Specific Prediction of Ligand Affinity with Structure-Based Interaction Fingerprints
Leidner, F., Yılmaz, N. K., & Schiffer, C. A. (2019) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.9b00457 · Cited by 31
Raimondi, D., Orlando, G., Vranken, W., & Moreau, Y. (2019) · Scientific Reports · doi:10.1038/s41598-019-53324-w · Cited by 31
Romero‐Rivera, A., Corbella, M., Parracino, A., Patrick, W. M., & Kamerlin, S. C. L. (2022) · JACS Au · doi:10.1021/jacsau.2c00063 · Cited by 31
Domain-PFP allows protein function prediction using function-aware domain embedding representations
Ibtehaz, N., Kagaya, Y., & Kihara, D. (2023) · Communications Biology · doi:10.1038/s42003-023-05476-9 · Cited by 31
Xia, C., Feng, S.-H., Xia, Y., Pan, X., & Shen, H.-B. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbac603 · Cited by 30
SURFMAP: A Software for Mapping in Two Dimensions Protein Surface Features
Schweke, H., Mucchielli, M.-H., Chevrollier, N., Gosset, S., & Lopes, A. (2022) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.1c01269 · Cited by 30
SE(3)-Stochastic Flow Matching for Protein Backbone GenerationPreprint
Bose, A. J., Akhound-Sadegh, T., Huguet, G., Fatras, K., Rector-Brooks, J., Liu, C., Nica, A. C., Korablyov, M., Bronstein, M. M., & Tong, A. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.02391 · Cited by 30
Ainsley, J., Mulholland, A. J., Black, G. W., Sparagano, O. A. E., Christov, C. Z., & Karabencheva‐Christova, T. G. (2018) · ACS Omega · doi:10.1021/acsomega.8b00385 · Cited by 29
A primer on model-guided exploration of fitness landscapes for biological sequence designPreprint
Sinai, S., & Kelsic, E. D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.10614 · Cited by 29
Lu, C., Lubin, J. H., Sarma, V., Stentz, S. Z., Wang, G., Wang, S., & Khare, S. D. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2303590120 · Cited by 28
Solid-state CD and peptide helical screw sense
Formaggio, F., Crisma, M., Toniolo, C., & Kamphuis, J. (1998) · Biopolymers · doi:10.1002/(sici)1097-0282(199603)38:3 · Cited by 27
TT3D: Leveraging precomputed protein 3D sequence models to predict protein–protein interactions
Sledzieski, S., Devkota, K., Singh, R., Cowen, L., & Berger, B. (2023) · Bioinformatics · doi:10.1093/bioinformatics/btad663 · Cited by 27
Ran, X., Jiang, Y., Shao, Q., & Yang, Z. J. (2023) · Chemical Science · doi:10.1039/d3sc02752j · Cited by 25
Dou, Z., Chen, X., Zhu, L., Zheng, X., Chen, X., Xue, J., Niwayama, S., Ni, Y., & Xu, G. (2024) · Nature Communications · doi:10.1038/s41467-024-53191-8 · Cited by 25
Das, S. K., Raucci, U., Neves, R. P. P., Ramos, M. J., & Parrinello, M. (2023) · ACS Catalysis · doi:10.1021/acscatal.3c01473 · Cited by 24
Witek, J., Podlewska, S., Rataj, K., Mordalski, S., & Bojarski, A. J. (2013) · Bioorganic & Medicinal Chemistry Letters · doi:10.1016/j.bmcl.2013.12.017 · Cited by 23
Learning Strategies in Protein Directed Evolution
Cadet, X. F., Gelly, J., van Noord, A., Cadet, F., & Acevedo‐Rocha, C. G. (2022) · Methods in molecular biology · doi:10.1007/978-1-0716-2152-3_15 · Cited by 22
Self-citationAllosteric regulatory control in dihydrofolate reductase is revealed by dynamic asymmetry
Kazan, I. C., Mills, J. H., & Ozkan, S. B. (2023) · Protein Science · doi:10.1002/pro.4700 · Cited by 22
Li, J., Guan, X., Zhang, O., Sun, K., Wang, Y., Bagni, D., & Head-Gordon, T. L. (2023) · PubMed · doi:10.48550/arxiv.2308.09639 · Cited by 22
Yang, L., Yang, G., Chen, X., Yang, Q., Yao, X., Bing, Z., Niu, Y., Huang, L., & Yang, L. (2021) · ACS Chemical Neuroscience · doi:10.1021/acschemneuro.1c00110 · Cited by 21
Goblirsch, B. R., Jensen, M. R., Mohamed, F. A., Wackett, L. P., & Wilmot, C. M. (2016) · Journal of Biological Chemistry · doi:10.1074/jbc.m116.760892 · Cited by 21
Lipase A from Bacillus subtilis: Substrate Binding, Conformational Dynamics, and Signatures of a Lid
Behera, S., & Balasubramanian, S. (2023) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.3c01681 · Cited by 21
Karlov, D. S., Long, S. L., Zeng, X., Xu, F., Lal, K., Cao, L., Hayoun, K., Lin, J., Joyce, S. A., & Tikhonova, I. G. (2023) · Structure · doi:10.1016/j.str.2023.02.014 · Cited by 17
Waksman, T., Astin, E., Fisher, S. R., Hunter, W. N., & Bos, J. I. B. (2024) · Molecular Plant-Microbe Interactions · doi:10.1094/mpmi-10-23-0154-fi · Cited by 15
Harding-Larsen, D., Madsen, C. D., Teze, D., Kittilä, T., Langhorn, M. R., Gharabli, H., Hobusch, M., Otalvaro, F. M., Kırtel, O., Bidart, G. N., et al. (2024) · ACS Omega · doi:10.1021/acsomega.4c01583 · Cited by 14
Self-citationPrediction of Sites of Metabolism of CYP3A4 Substrates Utilizing Docking-Derived Geometric Features
Feng, Y., Gong, C., Zhu, J., Liu, G., Tang, Y., & Li, W. (2023) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.3c00549 · Cited by 13
Dynamic Modelling Reveals ‘Hotspots’ on the Pathway to Enzyme-Substrate Complex Formation
Gordon, S. E., Weber, D. K., Downton, M. T., Wagner, J., & Perugini, M. A. (2016) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1004811 · Cited by 12
Endowing Protein Language Models with Structural KnowledgePreprint
Chen, D., Hartout, P., Pellizzoni, P., Oliver, C., & Borgwardt, K. M. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2401.14819 · Cited by 12
Graph representation learning for structural proteomics.
Fasoulis, R., Παλιούρας, Γ., & Kavraki, L. E. (2021) · PubMed Central · doi:10.1042/etls20210225 · Cited by 8
Tian, J., Dong, X. W., Wu, T., Wen, P., Liu, X., Zhang, M., An, X., & Shi, D. (2023) · International Journal of Biological Macromolecules · doi:10.1016/j.ijbiomac.2023.128405 · Cited by 8
Curado-Carballada, C., Feixas, F., & Osuna, S. (2019) · Advanced Synthesis and Catalysis · doi:10.1002/adsc.201900158 · Cited by 7
Xu, Z.-Q. J., Wu, J., Song, Y. S., & Mahadevan, R. (2022) · PubMed · Cited by 6
Assessing the performance of protein regression modelsPreprint
Richard, M., Kæstel‐Hansen, J., Groth, P. M., Bartels, S., Salomon, J., Tian, P., Hatzakis, N. S., & Boomsma, W. (2023) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2023.06.18.545472 · Cited by 5
Hoffbauer, T., & Strodel, B. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.01.12.575432 · Cited by 5
Kohout, P., Vasina, M., Majerova, M., Novakova, V., Damborský, J., Bednář, D., Marek, M., Prokop, Z., & Mazurenko, S. (2023) · ChemRxiv · doi:10.26434/chemrxiv-2023-jcds7 · Cited by 3
Self-citationInternal-Coordinate Density Modelling of Protein Structure: Covariance MattersPreprint
Arts, M., Frellsen, J., & Boomsma, W. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2302.13711 · Cited by 2
CATH protein domain classification (version 4.2)
Sillitoe, I., Dawson, N. L., & Orengo, C. A. (2019) · Figshare · doi:10.5522/04/7937330 · Cited by 2
Research works printed but not linked: 2
- Bellman, 1966, Dynamic programming, Science (1979), 153, 34
- Wold, 2011, 65, 1814
Cited by
Sorted by citations
Machine Learning-Driven Enzyme Mining: Opportunities, Challenges, and Future Perspectives
Moorhoff, F., Zhang, Y., Qiu, S., Dong, W., Medina-Ortiz, D., Zhao, J. B., & Davari, M. D. (2025) · ACS Catalysis · doi:10.1021/acscatal.5c04814 · Cited by 42
Grigorakis, K., Ferousi, C., & Topakas, E. (2025) · Catalysts · doi:10.3390/catal15020147 · Cited by 39
Machine learning in predictive biocatalysis: A comparative review of methods and applications
Tripathi, N., Hérisson, J., & Faulon, J. (2025) · Biotechnology Advances · doi:10.1016/j.biotechadv.2025.108698 · Cited by 37
Advances in Machine Learning Models for Predicting Enzyme Kinetic Parameters
Malli, A., Vasyutyn, D., & Kim, J. R. (2025) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5c02428 · Cited by 19
Samara, M. N., & Harry, K. D. (2025) · BioMedInformatics · doi:10.3390/biomedinformatics5030034 · Cited by 16
Metadata
- Authors
- Harding-Larsen, D., Funk, J., Madsen, N. G., Gharabli, H., Acevedo-Rocha, C. G., Mazurenko, S., & Welner, D. H.
- Affiliations
- Technical University of Denmark (DK)
- Masaryk University (CZ)
- Corresponding author
- Welner, Ditte Hededam
- ORCID
- Venue
- Biotechnology Advances
- Publisher
- Elsevier BV
- Published
- 2024-12
- Volume
- 77
- Pages
- 108459
- ISSN
- 0734-9750 (ISSN-L)
- OpenAlex
- W4403064295
- Cited by
- 38
- References
- 278
- Citations by year
- 2026: 17 · 2025: 17 · 2024: 4
- FWCI
- 1.55
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Biotechnology Advances (publishedVersion, cc-by)
- ChemRxiv (acceptedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- Technical University of Denmark, DTU Orbit (Technical University of Denmark, DTU) (submittedVersion, cc-by-nc-nd)
- Funders
- Horizon 2020
- Novo Nordisk Foundation
- Ministry of Education Youth and Sports of the Czech Republic
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Enzyme Catalysis and Immobilization
- Keywords
- protein representation, biocatalysis, enzyme engineering, learned representations, protein dynamics, protein 3D structure, protein sequences, inductive bias, self-supervised deep learning, explainability
- MeSH terms
- Machine Learning, Enzymes, Proteins, Neural Networks, Computer, Biocatalysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
