Splice Site Prediction: Transferring Knowledge Across Organisms

Printer-friendly versionSend by email
Conference Proceedings (fully refereed)
Kazantzidis, S., Krithara, A. and Paliouras, G.
As more genomes are sequenced, there is an increasing need for automated gene prediction. One of the subproblems of the gene prediction, is the splice sites recognition. In eukaryotic genes, splice sites mark the boundaries between exons and introns. Even though, there are organisms which are well studied and their splice sites are known, there are plenty others which have not been studied well enough. In this work, we propose two transfer learning approaches for the splice site recognition problem, which take into account the knowledge we have from the well-studied organisms. We use different representations for the sequences such as the n-gram graph representation and a representation based on biological motifs. Furthermore, we study the case where more than one organisms are available for training and we incorporate information from the phylogenetic analysis between organisms. An extensive evaluation has taken place. The results indicate that the proposed representations and approaches are very promising.
Software and Knowledge Engineering Laboratory (SKEL)
Conference Short Name: 
Conference Full Name: 
8th International Conference on Bioinformatics Models, Methods and Algorithms
Conference Country: 
Conference City: 
Conference Date(s): 
Tue, 21/02/2017 - Thu, 23/02/2017
Conference Level: 

© 2018 - Institute of Informatics and Telecommunications | National Centre for Scientific Research "Demokritos"

Terms of Service and Privacy Policy