Frequent Subsplit Representation of Leaf-Labelled Trees

• Published in: Evolutionary Computation, Machine Learning and Data Mining in Bioinformatics Vol. 4973; pp. 95 - 105
• Main Authors: Koperwas, Jakub, Walczak, Krzysztof
• Format: Book Chapter
• Language: English
• Published: Germany Springer Berlin / Heidelberg 2008; Springer Berlin Heidelberg
• Series: Lecture Notes in Computer Science
• Subjects: frequent subsplits; Leaf-labelled trees; phylogenetic trees
• ISBN: 9783540787563; 3540787569
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-540-78757-0_9

In this paper we propose an innovative method of representing common knowledge in leaf-labelled trees as a set of frequent subsplits, together with its interpretation. Our technique is suitable for trees built on the same leafset as well as for trees where the leafset varies. The proposed solution has a very good interpretation, as it returns different, maximal sets of taxa that are connected with the same relations in the input trees. In contrast to other methods known in literature it does not necessarily result in one tree, but may result in a profile of trees, which are usually more resolved than the consensus trees.