RBF Neural Network based Human Genome TSS Identification
| Authors: | Jie Chen, Beijing Institute of Technology, China
Zhihong Peng, Beijing Institute of Technology, China
Lijun Cao, Beijing Institute of Technology, China
Tingting Gao, Beijing Institute of Technology, China |
|---|
| Topic: | 8.2 Modelling & Control of Biomedical Systems |
|---|
| Session: | Biomedical Engineering / Biomedical Signal Processing II |
|---|
| Keywords: | Promoter recognition, Human genome, Transcription start site, RBF neural network. |
|---|
Abstract
Identification of functional motifs in a DNA sequence is fundamentally a statistical pattern recognition problem. This paper introduces a new algorithm for recognition of functional transcription start sites (TSSs) in human genome sequences, in which RBF neural network is adopted, and an improved heuristical method for 5-tuple feature viable construction is proposed and is implemented in two RBFPromoter and ImpRBFPromoter packages developed in Visual C++ 6.0. The algorithm is evaluated on several different test sequences sets. Compared with several other promoter recognition programs, this algorithm is proved to be more flexible with stronger learning ability and higher accuracy.
