Challenge of Searching in Ultra-Large-Scale Chemical Libraries
パトコア
As the library of existing and hypothetical chemical substances continues to grow, reaching sizes of many orders of magnitude, traditional cheminformatics tools and hardware infrastructure become inadequate. Questions that seem simple yet are fundamentally important for drug design, such as "Is this compound included in the library?" and "What are the ten most similar compounds in this collection?" become challenging to answer within acceptable response times in these ultra-large spaces. This paper discusses the current challenges in searching large chemical libraries and explores approaches that could provide drug designers, medicinal chemists, and cheminformaticians—who are trying to explore far beyond chemical space to track the next optimal and novel synthetically accessible bioactive structures—with groundbreaking solutions at the "Mars helicopter" level.
Related Links