Identification of TSS in the human genome based on a RBF neural network
- 40 Downloads
The identification of functional motifs in a DNA sequence is fundamentally a statistical pattern recognition problem. This paper introduces a new algorithm for the recognition of functional transcription start sites (TSSs) in human genome sequences, in which a RBF neural network is adopted, and an improved heuristic method for a 5-tuple feature viable construction, is proposed and implemented in two RBFPromoter and ImpRBFPromoter packages developed in Visual C++ 6.0. The algorithm is evaluated on several different test sequence sets. Compared with several other promoter recognition programs, this algorithm is proved to be more flexible, with stronger learning ability and higher accuracy.
KeywordsPromoter recognition human genome transcription start site RBF neural network
Unable to display preview. Download preview PDF.
- James W. Fickett, Artemis G. Hatzigeorgiou. Eukaryotic Promoter Recognition. Genome Research, vol. 7, no. 9, pp. 861–878, 1997.Google Scholar
- Tao Jiang, Ying Xu, Michael Q. Zhang. Current Topics in Computational Molecular Biology. Tsinghua University press, Beijing, pp. 101–255, 2002.Google Scholar
- G. B. Hutchinson. The prediction of vertebrate promoter regions using differential hexamer frequency analysis. Bioinformatics, vol. 12, no, 5, pp. 391–398, 1996.Google Scholar
- Michael Q. Zhang. Identification of Human Gene Core Promoters in Silico, Genome Research, vol. 8, pp. 319–326, 1997.Google Scholar
- Zhaoqi Bian. Pattern recognition, Tsinghua University Press, second edit, Beijing, pp. 250–270, 2000.Google Scholar