Splice site prediction using artificial neural networks

Abstract

A system for utilizing an artificial neural network to predict splice sites in genes has been studied. The neural network uses a sliding window of nucleotides over a gene and predicts possible splice sites. Based on the neural network output, the exact location of the splice site is found using a curve fitting of a parabolic function. The splice site location is predicted without prior knowledge of any sensor signals, like 'GT' or 'GC' for the donor splice sites, or 'AG' for the acceptor splice sites. The neural network has been trained using backpropagation on a set of 16965 genes of the model plant Arabidopsis thaliana. The performance is then measured using a completely distinct gene set of 5000 genes, and verified at a set of 20 genes. The best measured performance on the verification data set of 20 genes, gives a sensitivity of 0.891, a specificity of 0.816 and a correlation coefficient of 0.552. © 2009 Springer Berlin Heidelberg.