Transformation Of Exon Scores In Gene Prediction

Gene prediction gets more and more attention in the study of the genome. Actually it is used to predict the DNA transcription, DNA transformation and RNA shearing exactly on the non-noted RNA sequence which envoy people know the protein-coding zone. The method using computer to predict extrons becomes more and more important especially more and more genome sequencing was completed.In the past more than ten years gene-finding programs have got great development. Generally speaking they can be divided into two kinds. One is to identify new gene on the non-noted genome sequence by comparison of their ho-mology. We call this method homology comparison. But this method is limited. It has been shown that probably 50% new gene has not similar homology gene in the gene database. Another kind of method is to describe genome sequence under the gene signal discriminating frame. This kind of method is used to identify the coding region and the non-coding region on the DNA sequence by statistics of the codons. Great success has been made in recent years in this kind of method. We call this method signal discrimination ,for example: Gencsan and HMA gene.Most gene-finding programs will show the extron score. But among them only few possess some kind of practical significance. That is the evaluation of confidence level of the extron predicted. Sanja Rogic ect discover extron value given by Gencsan and HMMgene can actually reflect the reliability degree of extron predicted to some extent. The extron value provided by the rest procedure is either overestimate the reliability degree or totally has none sense. This...