.png){kind=icon}
返回
Diversity-oriented multi-compound synthesis optimization
.png){kind=icon}
PRE
AI
DOI：10.1039/d3re00610g.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Generative chemistry, which uses computational approaches to explore large chemical spaces, has gained considerable popularity in identifying potential lead candidates for drug discovery. However, a challenge with these methods is the lack of consideration of the synthetic feasibility of the generated molecules. This challenge can be addressed using compound generation and virtual screening approaches in combination with computer-aided synthesis planning (CASP) tools. However, the resulting synthesis effort may still be too costly in practice. To overcome this challenge, we present a method to generate a comprehensive set of compounds that effectively cover the chemical space of interest with minimal synthesis effort. The concept of using CASP systems for multi-compound optimization has been discussed previously. The approach presented in this paper goes beyond this and supports an efficient exploration of the chemical space. The goal is to select a small set of candidates (e.g. 25-50) from a larger pool of e.g. 500 candidates that can be synthesized in a few steps, while ensuring high diversity and broad distribution in chemical space. In this paper, we present an approach that effectively achieves both goals. Development of an optimization algorithm to identify comprehensive sets of compounds that effectively cover the chemical space of interest with minimal synthesis effort.
Keyword:
NETWORKS
DESIGN