The Frey Lab develops techniques that use large scale datasets to derive predictive models of how genes and many other genomic features act in combination to produce genetic messages that control cellular activities. We have most recently focused on applying deep learning to model "cell variables" and understand human diseases (Science, 2014; Nature Biotechnology 2015; Proceedings of the IEEE, 2015). The group is led by Brendan J. Frey, who has appointments in Engineering and Medicine. If you are interested in joining the group, click here.

Latest News
11/2015: Two of the Top 10 papers in 2015 came from our lab, as selected by RECOMB Regulatory and Systems Genomics.
07/2015: Frey lab launches a startup company Deep Genomics to transform precision medicine, genetic testing, diagnostics and the development of therapies.

Current Research Highlight

Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning

Regulatory motif predictions from DeepBind

The August issue of Nature Biotechnology highlighted our latest work on the cover. We show that deep learning architecture can discover the binding specificities of DNA and RNA-binding proteins from raw sequences. By accurately modelling how proteins bind to DNA and RNA, we can predict how mutations affect binding --- a key building block for computational identification of disease-causing mutations.


Reference

Babak Alipanahi, Andrew Delong, Matthew T Weirauch and Brendan J Frey. Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nature Biotechnology doi:10.1038/nbt.3300. Published online July 27 2015. [DeepBind article] [DeepBind web page]