Precision in Predicting Catalytic Performance of Medium Chain Dehydrogenase: A Benchmark Data Set and Comparative Metrics

Medium-chain dehydrogenases/reductases are highly versatile oxidoreductases with broad industrial relevance but complex structure-function relationships that continue to challenge catalytic performance prediction. A Benchmark for Medium-Chain Dehydrogenases (MCDBench) is introduced as a rigorously curated benchmark dataset comprising 792 enzyme-substrate-cofactor combinations, including experimentally validated inactive cases obtained under standardized conditions. MCDBench spans 36 chemically diverse alcohol substrates, representative MDR subclasses, and multiple cofactor types, enabling systematic evaluation of model generalization. Benchmarking four representative activity-prediction models (DLKcat, UniKP, TurNuP, and CataPro) shows that correlations with experimental activities remain modest under standardized conditions, and that distinguishing active from inactive catalytic space remains a practical challenge across enzyme classes and substrate types. Analysis points to a positive-data skew inherited from widely used biochemical databases and the relative scarcity of validated negative samples as key factors shaping this behavior. By providing a standardized benchmark that integrates experimentally validated negative cases, MCDBench offers a unified experimental and computational framework to help the field measure and further improve the discrimination of active and inactive catalytic space for biocatalyst design.

Authors

Institutions

Publication Details

Journal
bioRxiv (Cold Spring Harbor Laboratory)
Published
2026-11-17
DOI
https://doi.org/10.64898/2026.11.13.766581
Primary Topic
Enzyme Catalysis and Immobilization
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Precision in Predicting Catalytic Performance of Medium Chain Dehydrogenase: A Benchmark Data Set and Comparative Metrics

Ye Ni, Xiangyuan Gu, Jieyu Zhou, Xiaoyi Lin
4 links
bioRxiv (Cold Spring Harbor Laboratory)
Enzyme Catalysis and Immobilization
preprint

Precision in Predicting Catalytic Performance of Medium Chain Dehydrogenase: A Benchmark Data Set and Comparative Metrics

Ye Ni, Xiangyuan Gu, Jieyu Zhou, Xiaoyi Lin
preprint en

Abstract

Medium-chain dehydrogenases/reductases are highly versatile oxidoreductases with broad industrial relevance but complex structure-function relationships that continue to challenge catalytic performance prediction. A Benchmark for Medium-Chain Dehydrogenases (MCDBench) is introduced as a rigorously curated benchmark dataset comprising 792 enzyme-substrate-cofactor combinations, including experimentally validated inactive cases obtained under standardized conditions. MCDBench spans 36 chemically diverse alcohol substrates, representative MDR subclasses, and multiple cofactor types, enabling systematic evaluation of model generalization. Benchmarking four representative activity-prediction models (DLKcat, UniKP, TurNuP, and CataPro) shows that correlations with experimental activities remain modest under standardized conditions, and that distinguishing active from inactive catalytic space remains a practical challenge across enzyme classes and substrate types. Analysis points to a positive-data skew inherited from widely used biochemical databases and the relative scarcity of validated negative samples as key factors shaping this behavior. By providing a standardized benchmark that integrates experimentally validated negative cases, MCDBench offers a unified experimental and computational framework to help the field measure and further improve the discrimination of active and inactive catalytic space for biocatalyst design.

bioRxiv (Cold Spring Harbor Laboratory)
Jiangnan University (CN)
National Natural Science Foundation of China, Government of Jiangsu Province
Enzyme Catalysis and Immobilization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon