The use of mass spectrometry data to search molecular sequence databases is a well-established method for protein identification. The technique can be extended to searching raw genomic sequences, providing experimental confirmation or correction of predicted coding sequences, and has the potential to identify novel genes and elucidate splicing patterns.