SPLICE-SITE PREDICTION IN ARABIDOPSIS-THALIANA PRE-MESSENGER-RNA BY COMBINING LOCAL AND GLOBAL SEQUENCE INFORMATION

Artificial neural networks have been combined with a rule based system to predict intron splice sites in the dicot plant Arabidopsis thalian a, A two step prediction scheme, where a global prediction of the codi ng potential regulates a cutoff level for a local prediction of splice sites, is refined by rules based on splice site confidence values, pr ediction scores, coding context and distances between potential splice sites. In this approach, the prediction of splice sites mutually affe ct each other in a non-local manner, The combined approach drastically reduces the large amount of false positive splice sites normally haun ting splice site prediction, An analysis of the errors made by the net works in the first step of the method revealed a previously unknown fe ature, a frequent T-tract prolongation containing cryptic acceptor sit es in the 5' end of exons, The method presented here has been compared with three other approaches, GeneFinder, GeneMark and Grail, Overall the method presented here is an order of magnitude better, We show tha t the new method is able to find a donor site in the coding sequence f or the jelly fish Green Fluorescent Protein, exactly at the position t hat was experimentally observed in A.thaliana transformants, Predictio ns for alternatively spliced genes are also presented, together with e xamples of genes from other dicots, monocots and algae, The method has been made available through electronic mail (NetPlantGene@cbs.dtu.dk) , or the WWW at http://www.cbs.dtu.dk/NetPlantGene.html