Kroll Alexander, Lercher Martin J
Institute for Computer Science and Department of Biology, Heinrich Heine University, D-40225, Düsseldorf, Germany.
Biol Methods Protoc. 2024 Aug 24;9(1):bpae061. doi: 10.1093/biomethods/bpae061. eCollection 2024.
The recently published DLKcat model, a deep learning approach for predicting enzyme turnover numbers ( ), claims to enable high-throughput predictions for metabolic enzymes from any organism and to capture changes for mutated enzymes. Here, we critically evaluate these claims. We show that for enzymes with <60% sequence identity to the training data DLKcat predictions become worse than simply assuming a constant average value for all reactions. Furthermore, DLKcat's ability to predict mutation effects is much weaker than implied, capturing none of the experimentally observed variation across mutants not included in the training data. These findings highlight significant limitations in DLKcat's generalizability and its practical utility for predicting values for novel enzyme families or mutants, which are crucial applications in fields such as metabolic modeling.
最近发布的DLKcat模型是一种用于预测酶周转数（ ）的深度学习方法，它声称能够对任何生物体的代谢酶进行高通量预测，并捕捉突变酶的变化。在此，我们对这些说法进行批判性评估。我们表明，对于与训练数据序列同一性小于60%的酶，DLKcat的预测比简单地假设所有反应的恒定平均值更差。此外，DLKcat预测突变效应的能力比所暗示的要弱得多，无法捕捉训练数据中未包含的突变体之间实验观察到的任何变化。这些发现突出了DLKcat在通用性以及预测新酶家族或突变体的 值方面的实际效用方面的重大局限性，而这些在代谢建模等领域是至关重要的应用。
