DESCRIPTION (provided by applicant): The career goal of the candidate is to become an independent academic scientist in bioinformatics and computational genomics with a research focus on developing bioinformatics methods for the integrated analysis of large functional genomic datasets toward better understanding of gene regulation involved in human diseases. This award will facilitate the candidate's transition into an independent investigator by providing training in two primary bioinformatics areas in the context of human diseases: 1) gene regulatory interactions and networks between transcription factors and miRNAs; and 2) computational approaches for integrating and co-analyzing genomic profiling datasets. The mentors, Drs. Rakesh Nagarajan, Timothy Ley, and Wen-Hsiung Li are ideally suited to guide this award because each mentor has expertise and experience in the field of the proposed research and a long history of training and mentoring scientists. The objectives of this career development plan are to acquire additional training in areas that will enhance and extend the scope of the candidate's research and to establish the foundation for a future independent research career and subsequent funding. A didactic program in bioinformatics, computational genomics, and cancer genomics will complement the intellectual environment provided by the mentors, the meetings with other collaborators on this project, and national conferences in the related fields. The proposed project is summarized below.
It is becoming increasingly evident that the majority of human diseases are the product of multi-step processes, each of which involves the complex interplay of a multitude of genes acting at different levels of the genetic program. Elucidation of these complex mechanisms is crucial for a complete understanding of the pathophysiology of human diseases, which may lead to the discovery of novel biomarkers and therapeutic agents for human diseases. To achieve this goal, bioinformatics tools have been developed to construct gene regulatory networks to model these complex mechanisms. However, several obstacles exist that prevent reliable and accurate gene regulatory network inference. First, most current methods for gene regulatory network inference were designed for simpler organisms and cannot be applied to human diseases. Second, most current methods focus on modeling transcriptional regulation, excluding important miRNA regulators. Third, no current methods are able to leverage and co-analyze multiple genomic profiling data, including gene expression, gene copy number change, and mutation profiling. This research project proposes to develop novel bioinformatics tools for transcription factor and miRNA gene regulatory network inference to address these issues. The use of the new tool is demonstrated by the analysis of leukemia and breast cancer datasets. The bioinformatics tools outlined in this proposal will advance current techniques for gene regulatory network modeling, and will allow biomedical researchers to leverage the large-scale genomic profiling datasets in order to comprehensively elucidate gene regulatory pathways involved in human diseases.