The flood of high-throughput sequencing data necessitates computationally efficient algorithms for assembling genome sequences, comparing genome sequences, constructing evolutionary histories of genome sequences, and performing statistical inference on genome sequences. The same high-throughput sequencing technologies also allow us to understand the function of the genome and how processes like transcription, replication, packaging, and repair are dynamically regulated. Research at Duke has produced state-of-the-art methods and models for elucidating protein-DNA interactions, chromatin architecture, and transcriptional regulatory networks.