Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 27 min ago
PreprintCC BY-NC-NDChemRxiv

Data-driven strategies for enzyme engineering: From machine learning to protein language models

  • Shaozhen Ding1,2
  • Peng Wu3,4
  • Shubo Li1,2
  • Yuxuan Yang1,2
  • Han Zhu1,2
  • Heng Yu5
  • Yu Tian1,2
  • Dachuan Zhang6,7,8
  • Qian-Nan Hu3,4
8 affiliations
  1. 1Department of Chemical and Biomolecular Engineering, School of Life Science and Technology
  2. 2Wuhan Polytechnic University
  3. 3School of Artificial Intelligence
  4. 4Wuhan University
  5. 5Hubei Jiangxia Laboratory
  6. 6Department of Food Science and Technology
  7. 7National University of Singapore
  8. 8Bezos Centre for Sustainable Protein, National University of Singapore, 2 Science Drive 2, Singapore, 117542
Open at sourcePDFDOICite

Abstract

Enzymes are key biocatalysts that drive diverse biological processes and industrial applications, offering sustainable alternatives to conventional chemical catalysis. However, the enormous enzyme sequence space and the limited functional annotation have created an urgent demand for computational approaches to accelerate enzyme discovery and engineering. Recent advances in computational methods are accelerating enzyme engineering from empirical trial and error toward predictive design. This review summarizes representative computational tools and predictive models in the enzyme engineering workflow, including (1) enzyme data resources, (2) functional enzyme discovery, (3) enzyme property prediction (e.g., kinetics, optimal temperature and PH, solubility, promiscuity, and specificity), (4) de novo enzyme design, and (5) computationally guided enzyme optimization. We further compare current computational strategies, discuss their strengths and limitations in different engineering scenarios, and highlight emerging trends toward foundation models, generative artificial intelligence, multimodal learning, and closed-loop design–build–test–learn platforms.

Related

  1. From conservation scores to language models: a feature engineering perspective on machine learning for genetic variant interpretation
    Briefings in bioinformaticsPeer-reviewedCC BYFull textMolecular biology & gene regulationEurope PMC
  2. Building from the ground up: augmentation strategies to improve protein language model performance on protein sets
    E. Stump, H. Dennison, P. Strobbia +2
    bioRxivPreprintCC BY-NC-NDFull textBiophysicsbioRxiv
  3. Multi-Peptide Prompting Enables In-Context Learning in Protein Language
    J. Almonte, M. T. Vu, A. Ahn +1
    bioRxivPreprintCC BY-NC-NDFull textMolecular biology & gene regulationbioRxiv
  4. Functional alignment of protein language models via reinforcement learning
    Nathaniel Blalock, Srinath Seshadri, Kensuke Nakamura +4
    Nature CommunicationsPeer-reviewedCC BY-NC-NDFull text17 citedMolecular biology & gene regulationOpenAlex
  5. Novel hybrid protein scaffold gap filling using weighted machine learning ensemble, beam search, and mass-constrained reranking
    Tahmid Enam Shrestha, Md. Manzurul Hasan, Md. Rafiqul Islam
    arXivPreprintarXiv licenceFull textBiophysicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{ding2026datadriven,
  title = {{Data-driven strategies for enzyme engineering: From machine learning to protein language models}},
  author = {Shaozhen Ding and Peng Wu and Shubo Li and Yuxuan Yang and Han Zhu and Heng Yu and Yu Tian and Dachuan Zhang and Qian-Nan Hu},
  howpublished = {ChemRxiv},
  year = {2026},
  month = sep,
  doi = {10.26434/chemrxiv.15008605/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008605/v1},
  note = {CC BY-NC-ND},
}
RIS
TY  - UNPB
TI  - Data-driven strategies for enzyme engineering: From machine learning to protein language models
AU  - Shaozhen Ding
AU  - Peng Wu
AU  - Shubo Li
AU  - Yuxuan Yang
AU  - Han Zhu
AU  - Heng Yu
AU  - Yu Tian
AU  - Dachuan Zhang
AU  - Qian-Nan Hu
PY  - 2026
DA  - 2026/09/09
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15008605/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008605/v1
AB  - Enzymes are key biocatalysts that drive diverse biological processes and industrial applications, offering sustainable alternatives to conventional chemical catalysis. However, the enormous enzyme sequence space and the limited functional annotation have created an urgent demand for computational approaches to accelerate enzyme discovery and engineering. Recent advances in computational methods are accelerating enzyme engineering from empirical trial and error toward predictive design. This review summarizes representative computational tools and predictive models in the enzyme engineering workflow, including (1) enzyme data resources, (2) functional enzyme discovery, (3) enzyme property prediction (e.g., kinetics, optimal temperature and PH, solubility, promiscuity, and specificity), (4) de novo enzyme design, and (5) computationally guided enzyme optimization. We further compare current computational strategies, discuss their strengths and limitations in different engineering scenarios, and highlight emerging trends toward foundation models, generative artificial intelligence, multimodal learning, and closed-loop design–build–test–learn platforms.
ER  -