I’ve been thinking about how and when it is justified to use functional genomics data (e.g., histone modifications, DNA methylation, DNaseI hypersensitivity, etc.) collected in one cellular context for an analysis in a different context.  I performed a case study of this question for H3K27ac and H3K4me1 marks across developmental stages, tissues, and species. Check out the manuscript for the results:

Extrapolating histone marks across developmental stages, tissues, and species: an enhancer prediction case study [bioRxiv]

Congratulations to Vir Patel for being selected as a semifinalist in the 2014 Siemens Competition in Math, Science & Technology for his work with our group this summer. See the press release for more information.

We have been collaborating with Stephen Altschul’s group at the NIH to develop a more statistically sound way to represent DNA and protein sequence patterns using sequence logos. The paper describing our results just appeared in Bioinformatics:
Log-Odds Sequence Logos

If you’ve got some sequences you want to summarize, we highly recommend that you use the new tool (both web and command line versions are available):
LogOddsLogo