We are part of the Department of Data Science at Dana Farber Cancer Institute, the Department of Biostatistics at the Harvard T.H. Chan School of Public Health, and the Broad Institute.

We develop and apply computational methods to learn how spatial organization in tissues is related to normal and disease cell states. We study organization at multiple scales ranging from the tissue level architecture that governs cell-cell interactions, to the packaging of euchromatin and heterochromatin in the nucleus, down to the individual DNA loops that connect genes to their regulatory elements.

Tissue architecture

In addition to cell-intrinsic factors such as DNA mutations, many aspects of cell state are driven by external cues, determined in turn by a cell’s environment and the behaviors of its neighbors. We develop analysis methods for spatially resolved assays such as spatial transcriptomics and chromatin imaging to investigate these relationships in a range of solid tumors.

Nuclear organization

Cancer cells repackage their chromatin to such an extent that the microscopically visible changes form the basis for diagnosis of disease. Despite this, the underlying molecular changes remain largely unclear. We are interested in using imaging and 3-D genome mapping techniques to link diagnostic nuclear morphologies to gene activity and to understand their relationship to cell state and disease progression.

Regulatory DNA loops

Chromatin loops that bring linearly distant loci into close physical proximity are essential for much transcriptional regulation, and can either promote or inhibit gene activation in a context dependent manner. We have previously developed tools for identification and statistical testing of DNA loops, and are now focused on characterizing dynamic, transient regulatory interactions that appear, somewhat surprisingly, to represent the most common state of gene promoters.