Vaschetto Mariana, Weissbrod Tina, Bodle Dorothy, Güner Osman
Accelrys Ltd, 334 Science Park, Cambridge, CB4 0WN, UK.
Curr Opin Drug Discov Devel. 2003 May;6(3):377-83.
During the past few years, the introduction of ultra-high-throughput screening and new assay design and detection technologies has exponentially increased the amount and complexity of screening data. Effective use of this data implies a process that begins with assay design. An effective data management system should control a range of processes, from the initial selection of compounds and storage and mining of the assay result to more complex tasks, such as extracting patterns from these data. Remarkable advances have been made during the last year to increase efficiency at different phases of the screening, shifting the bottleneck of this process to data analysis. The challenge facing drug discovery today is to extract knowledge from these data. Knowledge discovery is defined as 'the non-trivial extraction of implicit, unknown, and potentially useful information from data'. A large amount of research is being devoted to optimize the extraction of knowledge from screening data. In this review, we discuss the screening process and its progress during the last year. Some of the challenges for the future, such as optimization of the knowledge discovery process and the sharing of data across an organization, will also be presented.
在过去几年中，超高通量筛选以及新的检测设计和检测技术的引入，使得筛选数据的数量和复杂性呈指数级增长。有效利用这些数据意味着要从检测设计开始。一个有效的数据管理系统应控制一系列流程，从化合物的最初筛选、检测结果的存储与挖掘，到更复杂的任务，比如从这些数据中提取模式。在过去一年里，为提高筛选不同阶段的效率取得了显著进展，将这一过程的瓶颈转移到了数据分析上。当今药物研发面临的挑战是从这些数据中提取知识。知识发现被定义为“从数据中提取隐含的、未知的且可能有用的信息这一不平凡的过程”。大量研究致力于优化从筛选数据中提取知识的过程。在本综述中，我们讨论了筛选过程及其在过去一年中的进展。还将介绍未来面临的一些挑战，比如知识发现过程的优化以及整个组织内的数据共享。
