The intersection between cancer biology and RNA processing

We study many different diseases where errors in RNA processing play important roles in disease initiation, progression, and response to therapy. Much of our work focuses on recurrent mutations affecting RNA splicing factors. These genetic lesions occur in many disparate diseases, including myelodysplastic syndromes, acute myeloid leukemia, chronic lymphocytic leukemia, melanomas, and cancers of the lung, breast, pancreas, bladder, and liver (among others). More recently, we have studied how aberrations in antigen presentation can enable cancer immune evasion.

In the long term, we seek to identify new biological phenomena that inform the development of better therapies.

Learn more

RNA processing

One of the most exciting findings of modern biology is that the flow of genetic information within the cell—from DNA to RNA to (commonly) protein—is very complex. For example, a single gene can encode the information necessary to make not just one, but many distinct protein isoforms, through the process of alternative splicing. In an extreme example, the Dscam gene in fruit flies can produce tens of thousands of distinct protein isoforms.

This process of alternative splicing affects almost all human genes. The regulation of alternative splicing thereby enormously increases the complexity of the human genome; conversely, errors in alternative splicing can cause diseases ranging from muscular dystrophies to cancer.

RNA processing other than splicing also contributes to human genome complexity and human disease. For example, RNA quality-control pathways such as nonsense-mediated decay guard against cellular noise. Some of our past studies suggest that dysfunction of this pathway contributes to the pathogenesis of facioscapulohumeral muscular dystrophy.

Specific diseases

Because much of our research investigates DNA or RNA abnormalities that occur in a wide range of cancer types, most of our studies are relevant to many different diseases. Nonetheless, we do frequently focus on specific disease types and model systems, particularly for preclinical studies. These include (among others):

We are particularly interested in finding molecular phenomena and biological processes that link otherwise unrelated disorders.

Our tools

We use many different technologies in our research, including high-throughput genomics, mouse modeling, bioinformatics, molecular genetics, and biochemistry. A typical experiment might combine a CRISPR/Cas9-based screen to identify a molecular phenotype with molecular experiments and mouse modeling to test specific predictions and confirm disease relevance in vivo.


We work closely with clinicians and physician-scientists in order to discover new basic science while answering questions of clinical relevance. We are always interested in new collaborations; please don't hesitate to contact us.