To fully leverage the current era of omics data revolution and realize the promise of precision medicine in cancer, we must address several key challenges: how to effectively integrate model organism studies with human data, model tumor heterogeneity, and unravel the systems-level impact of post-transcriptional modifications on cancer progression and treatment.

I propose to address these challenges through the development of sophisticated computational approaches, taking advantage of both the large quantity of publicly available cancer data as well as high quality data generated from cutting edge technologies (e.g., single-cell assays) through targeted collaborations. To address the challeng...

Read More

To fully leverage the current era of omics data revolution and realize the promise of precision medicine in cancer, we must address several key challenges: how to effectively integrate model organism studies with human data, model tumor heterogeneity, and unravel the systems-level impact of post-transcriptional modifications on cancer progression and treatment.

I propose to address these challenges through the development of sophisticated computational approaches, taking advantage of both the large quantity of publicly available cancer data as well as high quality data generated from cutting edge technologies (e.g., single-cell assays) through targeted collaborations. To address the challenges of translational research using model organisms, we will develop an integrative computational-experimental framework that automatically prioritizes the most relevant model organism, tissue context, and candidate disease genes for a particular cancer type (or subtype) of interest (Aim 1).

To model patient tumor heterogeneity, we will build cell-specific network models of cell types present in a patient tumor that will help us interpret and understand the dysregulated pathways in each cell type, and furthermore, allow us to prioritize targeted therapeutics based on the large catalogs of drug sensitivity profiles in cancer cell lines (Aim 2).

Finally, we will develop a network-based algorithm to enable modeling of cancer dysregulations in isoform interactions, allowing us to decipher another layer of complexity (on top of gene expression) in cancer progression and therapeutic response from a systems level perspective (Aim 3).

All methods and predictions will be made available to the wider biomedical community through open source software and dynamic, interactive visualizations. Together, the development and sharing of these methods will help us model complex mechanisms underlying cancer across different patients and propose more effective therapeutic regimens.

Read Less