Abstract
Identification of regions with untypical percentage G+C composition and dinucleotide signatures are two genome analysis techniques used in the identification of horizontally acquired DNA. We describe a generic program framework for performing both types of analysis in linear time. The approach is extended for length >2 oligonucleotide signatures. Using the derived program we test some of the conclusions of Karlin and Burge, the primary exponents of these techniques. We demonstrate that no single method of signature analysis is sufficient for the complete identification of horizontally acquired DNA. Consequently we produce a fast program - and a robust methodology - for the production and employment genome signatures in the identification of horizontally acquired DNA.