DNA sequences are key elements for both identification and classification of living organisms. Mainly for
historical reasons, a limited number of genes are currently used for this purpose. From a mathematical point of view, any
DNA segment, at any location, even outside of coding regions and even if they do not align, could be used as long as PCR
primers could be designed to amplify them. This paper describes two methods to search genomic data for the most
efficient DNA segments that can be used for identification and classification.