arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.07025cs.LGcs.AIq-bio.QMstat.ML
收藏

熵增强规划与大型语言模型用于药物发现

Entropy-Reinforced Planning with Large Language Models for Drug Discovery

University of Chicago(芝加哥大学);Argonne National Laboratory(阿贡国家实验室)

Xuefeng Liu, Chih-chan Tien, Peng Ding, Songhao Jiang, Rick L. Stevens

AI总结:

提出ERP算法,通过熵增强规划平衡探索与利用,提升Transformer解码在药物分子生成中的有效性,在多个基准上超越现有方法。

摘要:

药物发现的目标是识别具有针对结合靶点的特定药理学特性的化学化合物。现有的大型语言模型(LLMS)在分子生成的似然性方面可以达到较高的令牌匹配分数。然而,仅依赖LLM解码往往会导致生成无效的分子(由于单个误用的令牌)或次优的分子(由于LLM先验经验导致的探索与利用不平衡)。在此,我们提出ERP,即用于Transformer解码的熵增强规划,它采用熵增强规划算法来增强Transformer解码过程,并在利用与探索之间取得平衡。ERP旨在相比直接从Transformer采样,在多个属性上实现改进。我们在SARS-CoV-2病毒(3CLPro)和人类癌细胞靶蛋白(RTCB)基准上评估了ERP,并证明在这两个基准中,ERP始终比当前最先进的算法高出1-5个百分点,比基线高出5-10个百分点。此外,这种改进在不同目标训练的Transformer模型中是稳健的。最后,为了进一步说明ERP的能力,我们在三个代码生成基准上测试了我们的算法,并同样超越了当前最先进的方法。我们的代码公开于:https://github.com/xuefeng-cs/ERP。

英文摘要:

The objective of drug discovery is to identify chemical compounds that possess specific pharmaceutical properties toward a binding target. Existing large language models (LLMS) can achieve high token matching scores in terms of likelihood for molecule generation. However, relying solely on LLM decoding often results in the generation of molecules that are either invalid due to a single misused token, or suboptimal due to unbalanced exploration and exploitation as a consequence of the LLMs prior experience. Here we propose ERP, Entropy-Reinforced Planning for Transformer Decoding, which employs an entropy-reinforced planning algorithm to enhance the Transformer decoding process and strike a balance between exploitation and exploration. ERP aims to achieve improvements in multiple properties compared to direct sampling from the Transformer. We evaluated ERP on the SARS-CoV-2 virus (3CLPro) and human cancer cell target protein (RTCB) benchmarks and demonstrated that, in both benchmarks, ERP consistently outperforms the current state-of-the-art algorithm by 1-5 percent, and baselines by 5-10 percent, respectively. Moreover, such improvement is robust across Transformer models trained with different objectives. Finally, to further illustrate the capabilities of ERP, we tested our algorithm on three code generation benchmarks and outperformed the current state-of-the-art approach as well. Our code is publicly available at: https://github.com/xuefeng-cs/ERP.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑