Splice Site Detection in DNA Sequences using Probabilistic Neural Network

Abstract

With numerous of genomes sequenced, gene prediction has become a challenging problem in bioinformatics. Gene prediction helps in identifying physical and mental features of different organisms. A large number of gene prediction tools have been developed in the past two decades. Splice site detection method lies at the heart of ab-initio gene prediction tools and plays an important role in detecting the exon boundaries. In this paper, a method for detecting splice sites by using generalized regression neural network is proposed. The proposed method uses conditional probabilities to preprocess the input which enables it to incorporate the already known sequence features from biological knowledge. The experimental results show that the application of this new architecture to splice site detection has greatly improved the training time and reduces the false positive predictions.

Evaluating the Accuracy of Splice Site Prediction based on Integrating Jensen-Shannon Divergence and a Polynomial Equation of Order 2

“…Beside of PWM, neural network techniques have been widely applied in splice site detection methods. These methods use the complex non-linear transformation and learn the complex features of locality surrounding of the consensus AG/GT dinucleotides [7], [8]. Support vector machine is another method for splice site prediction [3].…”