Deciphering the genomic regulatory code that governs the generation of transcripts under different cellular conditions is one of the most important problems in biology. It would enable us to understand how the transcriptional landscape responds to changes in genomic features, eg, due to evolution or disease mutations. Inferring the regulatory code is difficult because of the combinatorial complexity of regulatory programs, an exponentially large search space, a low ratio of mRNA to DNA data, and poor modeling of cellular conditions. In this talk, I’ll describe my group’s efforts to infer regulatory codes that elucidate biological mechanisms, predict the behavior of new genes, and identify, in a much more powerful way than existing methods, the genetic determinants of human disease.