Weill Cornell Medicine
Hartman Institute for Therapeutic Organ Regeneration

Redmond Lab

David Redmond, PhD · Assistant Professor of Computational Biology

We work at the interface of computational and wet lab biology. We run the experiments and we build the methods, and neither half is a service function for the other.

We use single-cell and spatial measurement, across transcriptome, chromatin, protein and tissue, together with functional assays at the bench, to study how transcriptional regulation sets cell identity and how immune cells and the stroma around them shape each other, in development, inflammation, ageing and regeneration.

In practice that means building atlases of cell state, working out which regulators actually set identity, and testing whether they are enough to move a cell from one state to another. We perturb systems rather than just watching them, and we check a conclusion from more than one direction, at the bench and against independent human data. A result that only holds up in cross-validation is not really a result.

The systems are mostly vascular and immune. Endothelium is organ specialised, so a blood vessel in the pancreas, the liver, the bone marrow and the gut is a different cell doing a different job in each, and much of our work is on what makes it different and what it instructs its neighbours to do. That means working equally on the cells it talks to: haematopoietic stem cells in the marrow niche, macrophages and monocytes in tissue, and the leukocytes that arrive through the vessel wall.

Ageing and regeneration are where we push hardest, because they are the same question asked from two directions. Ageing shows you a niche in the act of failing, where tissue maintenance turns into tissue damage. Regeneration asks what would have to be put back to stop it. Disease is where the two meet, and where the answer has to be right.

Method development runs underneath all of it. We build the tools the questions need and release them, from one of the earliest pipelines for recovering paired full-length T-cell receptor sequences out of single-cell data to deep-learning models that read disease state directly from tissue images. Most of the work is done jointly with experimental groups whose systems put real pressure on the analysis, and the traffic runs both ways.