Minisatellites are genomic sequences comprised of short monomers which are successively repeated many times in the same direction. They are highly variable sequences in terms of their monomer building blocks, number of repeats, and their location on the genomic sequences. This mutability has made them excellent genetic markers for both linkage analysis and forensic science. This paper presents an accurate and highly efficient computer method for identifying all minisatellites in a given DNA, gene, or genomic sequence. It is based on a new indexing method which is intelligently used in such a way that does not use any main memory or secondary storage for storing search-keys. Furthermore, for each search-key value, which is not stored, a pointer points to a list occurrences of the search-key value in the sequence. Potential minisatellites are detected from these lists and actual ones are recognized. The software is capable of discovering all minisatellites of very large sequences such as human genome with 3.2 Giga base pairs in a very short time. The minimum number of repeats of the motif is set to be 3. An advantage of the software is in the detection of overlapped minisatellites that some state of the art software cannot detect. With respect to the time, looking for minisatellites with the proposed approach would be faster than TRF, Mreps and Dot2dot.

Keywords