We are broadly interested in the process intestinal epithelial cell specification. This process begins with the divison of stem cells that reside in the crypts of Lieberkuhn. The progenies of stem cells migrate up the intestinal gland, during which they encounter different environments and receive instructive cues to differentiate into one of the six functional intestinal cell types. Although individual pathways have been shown to modulate this process, cells in the physiological setting are simultaneously exposed to multiple factors resulting in multi-pathway activation. How cells integrate multivariate information to make cell decisions is currently undefined in disease contexts.
We utilize two technologies: highly multiplexable immunofluorescent microscopy (MultiOmyx) and mass cytometry (CyTOF), to generate quantitative signaling datasets amenable to mathematical modeling. Both of these technologies can detect and quantify a network-scale of proteins analytes at single cell resolution. These methods have their strengths and drawbacks. For example, MultiOmyx maintains localization information while CyTOF is capable of sampling the whole tissue. These techniques are complementary to obtaining rich datasets for data-driven modeling. We use mechanistic models to describe and predict functional interactions between measured pathways at the molecular level, while we use statistical modeling to integrate changes over time and space to arrive at phenotypic predictions.
We are applying our approaches to study how aberrant cellular signaling networks contribute to epithelial dysfunction in two complex intestinal diseases: inflammatory bowel disease and colorectal cancer. We hope that by obtaining a single-cell, systems-level understanding of pathological processes in diseases, we will devise more effective personalized therapies for these complex conditions.