New Methods for Cancer Class Discovery and Prediction: Integration, visualization

描述(由申请人提供)： 这项建议探索了新的计算方法，用于整合、分析和可视化癌症基因组图谱(TCGA)中快速增长的基因组和表观基因组信息。从长远来看，这些方法及其变体将能够严格识别分子生物标记物，以区分癌症、其亚型、其分期和结果，为开发改进的诊断和预后提供基础。它们还将能够识别对肿瘤的发生和发展至关重要的途径和过程，从而为治疗靶点的选择提供信息。到目前为止，大多数发现与癌症相关的类别差异的方法都是基于对mRNA转录的分析。在这里，我们探索先进的统计方法的修改、使用和调整，以整合TCGA数据，并使用我们的VISANT挖掘工具将TCGA与其他公开可用的数据整合。长期目标是开发将被广泛传播并用于发现癌症发展和进展的可靠生物标记物的方法，并对转化过程中发生的关键变化有更深入的了解。

DESCRIPTION (provided by applicant): This proposal explores new computational methods for integrating, analyzing and visualizing the rapidly growing genomic and epigenomic information in The Cancer Genome Atlas (TCGA). In the long range these methods and their variants will enable rigorous identification of molecular biomarkers for distinguishing cancer, their subtypes, theirs stages and their outcome, providing the basis for developing improved diagnostics and prognostics. They will also enable identification of the pathways and processes that are central to the initiation and progression of tumors, and thereby inform the choice of therapeutic target selection. Until now most methods for discovering class differences related to cancer have been based on the analysis of mRNA transcription. Here we explore the modification, use and adaptation of advanced statistical methods for integrating TCGA data, and the use of our VISANT mining tool for integrating TCGA with other publicly available data. The long term objective is to develop methods that will be widely disseminated and used to discover reliable biomarkers for cancer development and progression, and to gain a deeper understanding of the key alterations that occur during transformation.