MacMillan: Project 3; Technology Development and Research Project #3: Bioinformatics (White)

摘要：技术和研究开发项目#3：生物信息学负责 开发工具和方法来分析来自其他技术项目的数据，这些项目是 这项提议。这利用了UTSW和UCSC的数据分析和信息学优势来帮助 制作大型天然产物和植物药物库的广泛的分子和化学注释。 我们的交付成果将是数据和数据分析管道，它将帮助定义细胞学签名，以 化学实体，将化学实体映射到它们中断的信号通路，帮助识别 天然产品和植物药物所针对的正常和患病的生物网络，并有助于表明 这些化合物潜在的治疗/饮食应用。这将通过使用一系列 数据分析方法包括1)天然产物组分和植物活性的聚类分析 亲和传播聚类(APC)算法。该算法可应用于生成的数据要素 细胞学图谱、无偏代谢组学和融合。2)使用欧几里得距离分布 为聚变中的所有化学和遗传微扰物生成一个相似矩阵，以通过 关联关系。1)开发基于系统发育树算法的直观数据可视化工具。 开发这些工具不仅是为了可视化由TRD#1和TRD#2生成的数据，而且是为了通过 其他大规模的研究自然产物作用机制的方法。的最终目标是 这笔赠款的生物信息学部分是为了推动关于生物活性的假说的产生 正在评估的化学品，并向科学界提供这一信息。

Abstract: Technology and Research Development Project #3: Bioinformatics is responsible for the development of tools and methods for the analysis of data from the other technology projects that are part of this proposal. This takes advantage of the data analytics and informatics strengths of UTSW and UCSC to help produce a broad-scale molecular and chemical annotation of large libraries of natural products and botanicals. Our deliverable will be a data and data analytics pipeline that will help define a cytological signature to chemical entities, map chemical entities to signaling pathways they interrupt, help identify principal nodes in normal and diseased biological networks that natural products and botanicals target, and help indicate potential therapeutic/dietary applications for these compounds. This will be accomplished by using a series of data analysis methods including 1) Clustering analysis of natural products fractions and botanical activity by the affinity propagation clustering (APC) algorithm. This algorithm can be applied to data features generated by cytological profiling, unbiased metabolomics and FUSION. 2) Use of Euclidean distance distribution to generate a similarity matrix for all of the chemical and genetic perturbagens in FUSION to obtain “guilt by association correlations. 1) Develop tools for intuitive data visualization based on phylogenetic tree algorithms. These tools will be developed for visualization of not only the data generated by TRD#1 and TRD#2, but by other large scale methods looking at natural product mechanisms of action. The ultimate goal of the bioinformatics component of this grant is to drive the generation of hypotheses on the biological activity of the chemical being evaluated and provide this information to the scientific community.