Identifying splice-junction sequences by hierarchical multiclassifier

Splice sites are locations in DNA which separate protein-coding regions (exons) from noncoding regions (introns). Recently, several works have approached the splice-junction problem by applying methods from the field of machine learning techniques. In this work, we propose a hierarchical multiclassifier (HM) architecture, whose results show a drastically error reduction with respect to the performance of methods proposed in the literature.