发表机构
Emory University; University of Oxford; Singapore Management University（埃默里大学; 牛津大学; 新加坡管理大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
RouteFlow通过连续路线潜在空间中的奖励引导流匹配和循环一致性机制，实现高效可合成分子设计，在16个任务中取得最佳样本效率与逆合成成功率。
AI 中文摘要
目标导向的分子设计已迅速发展，然而，设计出的分子中有相当大比例在实际中难以合成，限制了其现实世界的实用性。先前的可合成性感知方法要么将生成的分子投影回偏离预期目标的类似物，要么直接在缺乏连续景观以进行高效搜索的离散合成空间中优化。我们认为这一局限性主要源于搜索空间而非优化器。为解决此问题，我们提出RouteFlow，一个将可合成分子设计重构为在连续路线潜在空间中进行搜索的框架，其中每个潜在向量映射回一条完整的合成路线，且可合成性被固有地保留。为导航此空间，我们采用奖励引导的流匹配作为高效采样器，引导向高属性区域移动。由于奖励优化可能将潜在向量推离真实合成路线的流形，导致解码不可靠，我们进一步引入循环一致性机制以稳定微调。在来自治疗数据共享库的16个优化任务中，RouteFlow在可合成性感知基线中实现了最佳样本效率，具有最佳合成可及性和最高逆合成成功率。我们的结果还证实，所提出的循环一致性可靠地保持优化在流形上，同时改善目标属性，支持有效的可合成分子发现。
英文摘要
Goal-directed molecular design has advanced rapidly, yet a substantial proportion of designed molecules remain difficult to synthesize in practice, limiting their real-world utility. Prior synthesizability-aware methods either project generated molecules back to synthesizable analogs that deviate from the intended target, or optimize directly in discrete synthesis spaces that lack a continuous landscape for efficient search. We argue that this limitation mainly comes from the search space rather than the optimizer. To address this, we propose RouteFlow, a framework that reformulates synthesizable molecular design as a search over a continuous route latent space, where each latent maps back to a complete synthesis route and synthesizability is inherently preserved. To navigate this space, we adopt reward-guided flow matching as an efficient sampler that steers toward high-property regions. Since reward optimization may push latents off the manifold of real synthesis routes, where decoding becomes unreliable, we further introduce a cycle-consistency mechanism to stabilize fine-tuning. Across 16 optimization tasks from Therapeutic Data Commons, RouteFlow achieves the best sample efficiency among synthesizability-aware baselines, with the best synthetic accessibility and the highest retrosynthesis success rate. Our results also confirm that the proposed cycle-consistency reliably keeps optimization on-manifold while improving target properties, supporting effective synthesizable molecular discovery.
