发表机构
Discovery Chemistry, Merck & Co., Inc.（默克公司发现化学部门）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出RxnCLF，一种基于凝聚反应图的自监督对比反应基础模型，经170万Pistachio反应预训练后，在多类产率预测基准上均优于基线模型，展现出良好泛化潜力。
AI 中文摘要
反应产率预测仍面临挑战，原因在于标注数据稀缺，且反应空间兼具组合规模庞大与稀疏分布的特点，这限制了现有反应表示的泛化能力。基于字符串、指纹和图的反应编码仅能部分捕捉化学变换，致使复杂底物反应的准确预测难度较大。我们提出反应对比学习基础模型（RxnCLF），这是一种用于反应表示学习的自监督对比框架。RxnCLF构建于凝聚反应图（CRG）之上，该图将反应物与产物信息统一为单张图，使模型能够学习显式且丰富的变换结构，而非分离的图结构。在170万个Pistachio反应上进行预训练后，RxnCLF学习到一个紧凑且连续的潜在空间，该空间既包含反应中心特征，又涵盖更广泛的侧链上下文，使其具备变换感知能力和化学可解释性。在多个产率预测基准上进行微调后，包括Buchwald-Hartwig、钯催化的BH偶联，以及专有HTE C-N偶联和酰胺形成数据集，RxnCLF始终优于基于图和序列的基线模型，提升了R²值并实现了整体最佳性能。我们的结果凸显了基于CRG的RxnCLF作为可扩展反应基础模型的潜力，其有望在更广泛的反应空间中泛化，并支持多种下游反应信息学任务，包括区域选择性预测、对映选择性预测以及反应条件优化。
英文摘要
Reaction yield prediction remains challenging because labeled data are scarce and reaction space is both combinatorially large and sparsely populated, limiting the generalization of existing reaction representations. String-, fingerprint-, and graph-based reaction encodings only partially capture chemical transformations, making accurate prediction difficult for reactions with complex substrates. We propose reaction contrastive learning foundation (RxnCLF), a self-supervised contrastive framework for reaction representation learning. RxnCLF is built on a condensed reaction graph (CRG) that unifies reactant and product information into a single graph, enabling the model to learn explicit and enriched transformation structure rather than disconnected graphs. Pretrained on 1.7 million Pistachio reactions, RxnCLF learns a compact and continuous latent space that captures both reaction-center features and broader side chain contexts, making it transformation-aware and chemically interpretable. Fine-tuned on multiple yield prediction benchmarks, including Buchwald-Hartwig, Pd-catalyzed BH coupling, and proprietary HTE C-N coupling and amide formation datasets, RxnCLF consistently outperforms graph and sequence-based baselines, improving R2 and achieving the best performance overall. Our results highlight the promise of CRG-based RxnCLF as a scalable reaction foundation model, with the potential to generalize across broader reaction spaces and support diverse downstream reaction informatics tasks, including regioselectivity prediction, enantioselectivity prediction, and reaction condition optimization.
Comments8 pages, 6 figures
