ChemMine Tools: an Open Source Framework for Chemical Genomics

加州大学河滨分校（University of California, Riverside）获得了开发ChemMineTools的资助，这是一个能够有效分析和模拟大量小分子及其生物活性数据的环境。它将提供对一组可扩展的开源工具的无限制访问，这些工具集成了新颖的和现有的算法。为了最大限度地提高实验和计算科学家的效用范围，分析模块将从强大的R环境中获得，以及直观易用的web界面。该项目的具体目标是：(1)开发和实现加速复合搜索和聚类算法，这些算法可扩展到当今拥有数百万条目的大型数据库。本文将重点介绍通过嵌入和索引（EI）对超快速EI- search和EI- clustering算法的扩展。这些多用途算法将被用于高级相似度度量，目前由于其速度性能不足而无法用于处理大型数据库。(2)开发R软件包ChemMine R Tools。它将提供高级集群、机器学习和可视化功能以及交互式可视化工具。(3) ChemMine Web tools界面将提供对所有分析和可视化工具的用户友好访问。(4)将提供一项教育外展计划，提供广泛的培训机会，并将代表性不足的群体纳入该项目。通过在有效的数据挖掘环境中整合新的和现有的分析程序，该项目将传播和转变现代生物学中强大化学方法的多学科概念。此外，它将鼓励年轻科学家将化学信息学策略纳入他们的日常研究中。本计划将提供计算生物学与化学信息学交叉的跨学科训练所需的大量教育资源。研讨会将提供给科学家，博士后研究人员，研究生，本科生，高中和其他K-12学生。代表性不足的群体成员将参与该项目的各个方面。将提供广泛的在线讲习班和软件手册，以最大限度地扩大这些活动的教育范围。有关该项目的更多信息可在其网站上找到：http://cmtools.ucr.edu。

The University of California, Riverside is awarded a grant to develop ChemMineTools, an environment to efficiently analyze and model large sets of small molecules along with their bioactivity data. It will provide unrestricted access to a scalable set of open source tools that integrates novel and existing algorithm. To maximize its utility spectrum for experimental and computational scientists, the analysis modules will be available from the powerful R environment, as well as an intuitive-to-use web interface. The specific objectives of this project are: (1) The development and implementation of accelerated compound search and clustering algorithms that scale to today's large databases with millions entries. This will focus on the expansion of the ultrafast EI-Search and EI-Clustering algorithms by embedding and indexing (EI). These multipurpose algorithms will be adopted to advanced similarity measures that can currently not be used for processing large databases due their insufficient speed performance. (2) The R package ChemMine R Tools will be developed. It will offer access to advanced clustering, machine learning and visualization functionalities along with interactive visualization tools. (3) User-friendly access to all analysis and visualization tools will be provided by the ChemMine Web Tools interface. (4) An educational outreach program will be offered to provide extensive training opportunities and to integrate underrepresented groups into this project. By integrating novel and existing analysis routines in an efficient data mining environment, the project will disseminate and transform multidisciplinary concepts of powerful chemical approaches in modern biology. Moreover, it will encourage young scientists to incorporate cheminformatics strategies into their daily research. Substantial educational resources for interdisciplinary training at the intersect of computational biology and cheminformatics will be provided by this project. Workshops will be offered to scientists, postdoctoral researchers, graduate, undergraduate, high school and other K-12 students. Members of underrepresented groups will participate in all aspects of this project. Extensive online workshop and software manuals will be provided to maximize the educational outreach of the activities. Further information about this project may be found at its website: http://cmtools.ucr.edu.