Novel Tools for the Vertical Analysis of Genomic Data

由于缺乏优秀的生物信息学专业人员，许多昂贵的基因组学实验仍然分析不足。该提案的目标是提供一套交钥匙生物信息学工具，该工具：i）可以处理来自当前高通量技术（如微阵列）的大量数据，以及ii）允许从组件级数据组装系统级信息的垂直推理。本文提出的工作将提供一个新的数据库和相关的数据挖掘工具，允许：（1）将基因表达谱转换为功能谱，显示在给定条件下各种细胞功能如何变化，并评估其统计学意义，（2）发现在给定条件下相互作用与正常相互作用显著不同的生物过程。 （3）确定在给定实验中最有可能与生物学相关的途径，并定量评估其影响。 （4）通过对照现有生物学知识检查其结果的一致性来验证高通量实验，例如微阵列 (5)计算机全身数据挖掘技术能够在短时间内提供任何给定基因在所有感兴趣组织中的表达谱。通过建立和维护一个具有先进数据挖掘能力的公共数据库来加强研究基础设施。2.交叉培训计算机科学和生物学研究人员。通过以下方式扩大代表性不足群体的参与：该项目的主要目的是：（a）为少数民族学生提供培训;（b）每年对特定的少数民族服务机构进行两次招聘访问;（c）为处境不利的K-12学生组织一次开放日;（d）有针对性地宣传通过该项目资助的职位。4.通过以下方式广泛传播科学知识：i）为该项目期间开发的工具提供网络接口，使科学界能够轻松访问这些工具; ii）组织一次年度多学科会议，将计算机科学家和生物学家聚集在一起。

Due to the scarcity of good bioinformatics professionals, many expensivegenomics experiments remain under-analyzed. The goal of this proposal is to provide a suite of turnkey bioinformatics tools that: i) can process extremely large amounts of data as those coming from current high-throughput technologies such as microarrays, and ii) allow vertical inferences assembling system level information from component level data. The work proposed here will provide a novel database and associated data mining tools allowing: (1)Translation of gene expression profiles into functional profiles showing how various cellular functions are changed in the given condition and assessing their statistical significance, (2) Discovering those biological processes whose interaction in a given condition is significantly different their normal interaction. (3)Identifying those pathways that are most likely to be biologically relevant in agiven experiment and quantitatively assessing their impact. (4)Validating high-throughput experiments, such as microarrays, by checking the consistency of their results against existing biological knowledge (5) In-silico whole-body data mining techniques able to provide in a short time an expression profile of any given gene across all tissues of interest.The broader impact aspects of this proposal include the following:1. Enhance the research infrastructure by constructing and maintaining a public database with advanced data mining capabilities. 2. Cross-train researchers in computer science and biology.3. Broaden the participation of under-represented groups by: i) recruiting Ph.D. students ii) undertaking two annual recruitment visits to specific minority-serving institutions; iii) organizing an open house for disadvantaged K-12 students; and iv) targeted advertising of the positions funded through this project. 4. Broadly disseminating scientific knowledge by: i) providing a web interface to the tools developed during this project allowing the scientific community to access them easily ; and ii)organizing an annual multidisciplinary conference bringing together computer scientists and biologists.