As a statistician and computational biologist, my research interest ranges from innovative methodology development to solving challenges in data analysis and statistical learning in biomedical studies. At ISMMS, I got the opportunities to participate in large scale cancer studies from NCI-CPTAC (Clinical Proteomic Tumor Analysis Consortium) and have been focusing my research on integrative proteo-genomic analysis methods and their application to complex high-dimensional multimodality cancer data sets. Specifically, I develop novel computational/statistical methods for integrating different types of -omics data to understand the molecular mechanism and identify signaling pathways driving the disease that, in turn, can be translated into the clinic. Specifically, involvement in CPTAC proteogenomic cancer studies motivated me to develop novel algorithms for learning directed network based on high dimensional omics data. I have recently developed a new tool to build high dimensional graphic model based on proteogenomic data to facilitate system-learning and to screen for protein/gene markers driving platinum response in ovarian cancer. We have also applied DAG models on spatial transcriptomics (ST) data to infer directed network between ligands and receptors. We have recently proposed a novel statistical method to characterize the ligand receptor interaction networks between adjacent tumor and stroma cells based on ST data.