Motivation: The large number of sequenced genomes required the development of software that reconstructs the consensus sequences of transposons and other repetitive elements. However, the available tools usually focus on the accurate identification of raw repeats and provide no information about the taxonomic position of the reconstructed consensi. TEclass is a tool to classify unknown transposable elements into their four main functional categories, which reflect their mode of transposition: DNA transposons, long terminal repeats (LTRs), long interspersed nuclear elements (LINEs) and short interspersed nuclear elements (SINEs). TEclass uses machine learning support vector machine (SVM) for classification based on oligomer frequencies. It achieves 90-97% accuracy in the classification of novel DNA and LTR repeats, and 75% for LINEs and SINEs.
Availability: http://www.compgen.uni-muenster.de/teclass, stand alone program upon request.