GP-Based Method for Extracting Exons from DNA Sequence

This paper describes GP-based method for extracting exons from DNA sequences. The advantage of GP method is automatically to build a proper model for identification. We made experiments to identify the edges of exon (exon-intron boundary and intron-exon boundary) in DNA sequences and compare the average identification rate by GP method with that by conventional methods (Genscan, HMMgene, Genie, FGENES, Morgan, GeneMark.hmm and MZEF) with data set 'HMR195'. As for exon-intron boundary, GP method is superior to all the conventional methods and as for intron-exon boundary, GP method is superior to almost all the conventional methods except Genscan. Key Words: Genetic Programming, Identification, Exon, Splice sit