Applications of resampling methods

本研究项目将着眼于基于重采样方法的三个应用领域。首先是对设计实验中产生的基因表达数据的分析，特别是对基因与环境相互作用的研究。这些相互作用被广泛假设存在于许多复杂疾病中。然而，在基因表达水平上发现这些相互作用并不容易。在这种情况下，标准参数测试通常不太有用，但可以使用基于排列和重新采样的方法。大多数微阵列分析没有考虑到单个个体基因之间的依赖性。然而，有很多关于基因功能和关系的信息。我将研究如何将这些信息合并到分析中。我还将研究在处理多重测试问题和估计错误发现率时考虑遗传信息的方法。

This research program will look at three areas of application for resampling-based methodologies. The first is to the analysis of gene expression data arising from designed experiments and in particular to the search for gene-environment interactions. These interactions are widely hypothesized to exist in many complex diseases. Finding these interactions at a gene-expression level, however has not been easy. Standard parametric tests are not generally useful in this setting but permutation and resampling based methods can be used. Most analyses of microarrays does not take into account the dependence between genes for a single individual. There is a lot of information about gene function and relationships available, however. I will examine how we can incorporate this information into the anlysis. I will also look at ways to take genetic information into account in dealing with the multiple testing problem and estimating the false discover rate.