Hamming-Clustering method for signal prediction in 5' and 3' regions of eukaryotic genes

Gene expression is regulated by different kinds of short nucleotide domains. These features can either activate or terminate the transcription process or can regulate the speed of the chemical reactions involved in this event. In order to predict the signal sites in the 5' and 3' gene regions we have applied the Hamming-Clustering network (HC) to the TATA-box, to transcription initiation site and to the poly(A) signal. The HC is a new model of Artificial Neural Network (ANN) specially designed for working with binary data. More then 1000 Poly-A signals have been extracted from EMBL database rel. 42 and used to build the training and the test set. A full set of the Eukaryotic genes (1252 entry) from the Eukaryotic Promoter Database (EPD rel. 42) have been used for the TATA-box signal and transcription initiation site training. A set of eukaryotic plant genes have been used to test the validity of the Hamming-Clustering network approach.