My graduate work in bioinformatics aimed to develop a framework that identifies tissue-specific enhancers by integrating multi-modal genomic (gene expression, methylation and interaction) data. As a postdoctoral fellow in Carnegie Mellon University, I obtained training in data mining methods for image analysis. Previously, at the MD Anderson Cancer Center, I developed methods to analyze the relationship between image-derived phenotypic attributes and genotypic attributes for cancer patients, focusing most of the work on datasets from the Cancer Genome Atlas (TCGA). Such image data is obtained from multiple modalities including, HandE whole slide data, immunohistochemistry, high-content microscopy, and radiology. At the University of Michigan, the primary focus of our group is to develop algorithms to quantify tumor phenotypes (from radiology, histopathology and microscopy data) and to relate those phenotypic aspects with genetic alterations like expression changes, mutation or copy number alterations. Further, we also develop methodologies (using statistical methods and machine learning) to integrate these diverse modalities (imaging, genomics etc.) to develop decision algorithms for prognosis and treatment selection. Looking forward, we are interested to develop analysis methodologies (based on spatial statistics, image analysis and machine learning) as well as tools for "spatial cancer systems biology" by leveraging spatial transcriptomics, spatial proteomics and multiplex imaging strategies of the tumr microenvironment to obtain a deeper characterization of regulatory networks underlying oncogenesis.