VISUAL AND ANALYTICAL TOOLS FOR CLUSTER ANALYSIS

本子项目是利用由NIH/NCRR资助的中心赠款提供的资源的众多研究子项目之一。子项目和研究者（PI）可能已经从另一个NIH来源获得了主要资金，因此可以在其他CRISP条目中表示。列出的机构是中心的，不一定是研究者的机构。该项目的目标是设计新的分析和可视化工具和技术，以提供对多个聚类算法结果的记录级信息的见解。我们的工作是通过设计工具来解决生物医学领域的重大问题，使科学家能够查看多种聚类方法并分析结果，以帮助数据探索。生物医学领域在过去十年中经历了强劲的增长，产生了大量需要新方法的数据集。这一领域创造了一种新的研究范式，该领域的研究方法不同于已有研究程序的既定科学领域。利用视觉技术来驾驭实验科学家的直觉似乎是常识，但对自动计算的强烈依赖仍然弥漫在这个领域。智力上的冒险是这项事业的一部分，因为在参与的学科中，实践的个人会求助于彼此之间的互动，试图说一种新的共同语言，并在一个与他们自己的领域相距甚远的领域取得突破。本研究将完成三个主要目标：制定、实施和验证多个聚类结果相似性的分析措施（1年级和2年级）：(a)正式定义聚类结果的相似性和差异性(b)创建捕获相似性和差异性定义的新分析措施(c)定义和实施分析措施的具体应用；具体目标2。构建新的可视化，帮助识别相似和不同的记录，集群和子集（3年级及以上）(a)设计和实施与新的分析措施一起工作的新颖可视化技术(b)创建和实施促进探索的互动方法；具体目标3。调查和改进技术和算法的性能，并在大型实验数据集上应用和扩展技术和工具（第5年）。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. The goal of this project is to design new analytical and visual tools and techniques to provide insights into the record level information of multiple clustering algorithm results. Our work is going to address significant questions in the biomedical field by devising tools that enable scientists to look through multiple clustering methods and analyze the results, in order to aid the data exploration. The biomedical field has undergone strong growth over the past decade, generating massive data sets that require new approaches. A new research paradigm is created in this area, and the fields' approaches differ from the established fields of science that have recognized research procedures. Using visual techniques in order to harness the experimental scientist's intuition seems common sense but the strong dependence on automatic computation still pervades the field. Intellectual risk-taking is part of this undertaking, since practicing individuals in participating disciplines would otherwise recourse to interact with one another, attempt to speak a new common language, and make breakthroughs in a field so distant from their own. This research will accomplish three primary goals: Specific Aim 1. Develop, implement and validate analytical measures of similarity of multiple clustering results (Year 1 and Year 2): (a) Formally define the similarity and dissimilarity of clustering results (b) Create novel analytical measures that capture the definitions of similarity and dissimilarity (c) Define and implement specific applications of the analytical measures; Specific Aim 2. Construct novel visualizations that aid the identification of similar and dissimilar records, clusters and subsets (Year 3 and beyond) (a) Design and implement novel visualization techniques that work with the new analytical measures (b) Create and implement interaction methods that facilitate the exploration; Specific Aim 3. Investigate and improve the performance of the techniques and algorithms and apply and extend the techniques and tools on large experimental data sets (Year 5).