A classification of alternatively spliced cassette exons using AdaBoost-based algorithm

Alternative splicing (AS) is a mechanism for generating different gene transcripts (called isoforms) from the same genomic sequence. Accurate classification of alternative splicing is important for understanding the mechanism of gene regulation. Although the different algorithms are applied to classify AS, the accuracy of these algorithms are still unsatisfactory. In this paper, we propose a new classification algorithm to classify the cassette exons and constitutive exons with existed and new features based on the AdaBoost algorithm. Experimental results show that the accuracy is higher than current algorithms. The sensitivity is 81.92% and the specificity is 74.06%, namely that 81.92% of cassette exons and 74.06% of constitutive exons were correctly classified.