Cross-Domain Descriptor Fusion: Integrating Biological and Materials Descriptors for Hybrid Catalyst Design

While most computational catalyst discovery has focused on either biological (enzyme) or inorganic catalyst classes separately, there’s a vast untapped opportunity in merging insights from both domains. Papers like Lauko et al. (2024, 2025) demonstrate deep learning for de novo enzyme design, while Zhai et al. (2022) and Hashimoto et al. (2025) focus on high-entropy and inorganic materials. My idea is to build a computational pipeline that extracts and fuses descriptors—such as active site polarity, flexibility, and hydrophobicity from enzymes, with coordination geometry, band structure, and electron density from inorganic catalysts. Feeding this rich, hybrid descriptor set into ML models could enable the design of “bio-inorganic hybrid catalysts” optimized for challenging transformations, especially those requiring both substrate specificity and robust turnover. This approach could open new frontiers for reactions that neither pure enzymes nor inorganics efficiently catalyze alone, and it represents a novel synthesis of methodologies across domains.

References:

1. High‐Entropy Catalyst—A Novel Platform for Electrochemical Water Splitting. Yiyue Zhai, Xiangrong Ren, Bolun Wang, S. Liu (2022). Advanced Functional Materials.
2. Computational design of serine hydrolases. Anna Lauko, S. Pellock, Ivan Anischanka, K. Sumida, David Juergens, Woody Ahern, Alex Shida, Andrew C Hunt, Indrek Kalvet, Christoffer H Norn, Ian R. Humphreys, C. Jamieson, A. Kang, Evans Brackenbrough, Asim K. Bera, Banumathi Sankaran, K. N. Houk, David Baker (2024). bioRxiv.
3. Computational design of serine hydrolases.. Anna Lauko, S. Pellock, K. Sumida, Ivan Anishchenko, David Juergens, Woody Ahern, Jihun Jeung, Alex Shida, Andrew C Hunt, Indrek Kalvet, Christoffer H Norn, Ian R. Humphreys, C. Jamieson, Rohith Krishna, Yakov Kipnis, A. Kang, Evans Brackenbrough, Asim K. Bera, Banumathi Sankaran, K. N. Houk, David Baker (2025). Science.
4. A materials map integrating experimental and computational data via graph-based machine learning for enhanced materials discovery. Yusuke Hashimoto, Xue Jia, Hao Li, Takaaki Tomai (2025). APL Machine Learning.