My doctoral dissertation comprises of three different projects that address three different aspects of statistical analysis of genomics data.

The first project describes a novel statistical approach for identification of 'master regulator' transcription factor in a genome. A 'master regulator' transcription factor, being at the top of the hierarchy of the transcriptomic regulation, may control the regulatory activities of the other transcription factors and the associated genes. For example, in cases of systemic disease, cellular function is disrupted. We hypothesize that these may be the result of a regulatory disruption in the most upstream elements of the regulatory cascade, the 'master regulator'. Therefore, it is important to identify and target the master regulator transcription factor for proper understanding of the associated disease process. Through simulated scenario and real dataset analyses, we show that our method performs well in validating the existence of a master regulator, and identifies biologically meaningful master regulators.

The second project involves an integrated analysis using multiple cancer datasets for investigating the significance of the biological pathways which are interrupted by cancerassociated genetic mutations. This dataset consists of expression profiles for genes/proteins of patients receiving treatment, for three types of cancer---Head and Neck Squamous Cell Carcinoma, Lung Adenocarcinoma and Kidney Renal Clear Cell Carcinoma. We consider pathway analysis to identify all the biological pathways which are active among the patients and investigate the roles of the significant pathways using a differential network analysis of the protein expression datasets for the three cancers separately. We then integrate the pathway based results of all the three cancers which provide a more comprehensive picture of the three cancers.

In the third project, we develop a novel meta-analysis method of combining p-values from different independent experiments involving large-scale multiple testing frameworks. Adhering to the regular statistical assumptions regarding the null distributions of test statistics can lead to incorrect significance testing results and biased inference in large-scale multiple testing frameworks when results from different independent genomic experiments are combined. In order to overcome this, our proposed method takes into account empirical adjustments of the individual test statistics and p-values. Consequently, our method outperforms the standard metaanalysis approach of significance testing as shown in simulation studies and real genomic data analyses.