Fast FPT Algorithms for Computing Rooted Agreement Forests: Theory and Experiments
We improve on earlier FPT algorithms for computing a rooted maximum agreement forest (MAF) or a maximum acyclic agreement forest (MAAF) of a pair of phylogenetic trees. Their sizes give the subtree-prune-and-regraft (SPR) distance and the hybridization number of the trees, respectively. We introduce new branching rules that reduce the running time of the algorithms from O(3 k n) and O(3 k n logn) to O(2.42 k n) and O(2.42 k n logn), respectively. In practice, the speed up may be much more than predicted by the worst-case analysis. We confirm this intuition experimentally by computing MAFs for simulated trees and trees inferred from protein sequence data. We show that our algorithm is orders of magnitude faster and can handle much larger trees and SPR distances than the best previous methods, treeSAT and sprdist.
KeywordsSearch Tree Recursive Call Protein Tree Canada Research Chair Agreement Forest
Unable to display preview. Download preview PDF.
- 6.Bordewich, M., Linz, S., John, K.S., Semple, C.: A reduction algorithm for computing the hybridization number of two trees. Evol. Bioinform. 3, 86–98 (2007)Google Scholar
- 12.Hickey, G., Dehne, F., Rau-Chaplin, A., Blouin, C.: SPR distance computation for unrooted trees. Evol. Bioinform. 4, 17–27 (2008)Google Scholar
- 13.Hillis, D.M., Moritz, C., Mable, B.K. (eds.): Molecular Systematics. Sinauer Associates (1996)Google Scholar
- 14.Maddison, W.P.: Gene trees in species trees. Syst. Biol. 46(3), 523–536 (1997)Google Scholar
- 17.Whidden, C., Beiko, R.G., Zeh, N.: Fast FPT algorithms for computing rooted agreement forests: Theory and experiments. Tech. Rep. CS-2010-03, Faculty of Computer Science, Dalhousie University (2010)Google Scholar
- 18.Whidden, C., Zeh, N.: A unifying view on approximation and FPT of agreement forests. In: Salzberg, S.L., Warnow, T. (eds.) WABI 2009. LNCS(LNBI), vol. 5724, pp. 390–401. Springer, Heidelberg (2009)Google Scholar
- 19.Whidden, C.: rSPR FPT Software, http://kiwi.cs.dal.ca/Software/RSPR