Researchers have succeeded in teaching computers how to identify commonalities in DNA sequences known to regulate gene activity, and to then use those commonalities to predict other regulatory regions throughout the genome. The tool is expected to help scientists better understand disease risk and cell development.