Results 1  10
of
13
A Simple, Fast, and Accurate Algorithm to Estimate Large Phylogenies by Maximum Likelihood
, 2003
"... The increase in the number of large data sets and the complexity of current probabilistic sequence evolution models necessitates fast and reliable phylogeny reconstruction methods. We describe a new approach, based on the maximumlikelihood principle, which clearly satisfies these requirements. The ..."
Abstract

Cited by 2182 (27 self)
 Add to MetaCart
The increase in the number of large data sets and the complexity of current probabilistic sequence evolution models necessitates fast and reliable phylogeny reconstruction methods. We describe a new approach, based on the maximumlikelihood principle, which clearly satisfies these requirements. The core of this method is a simple hillclimbing algorithm that adjusts tree topology and branch lengths simultaneously. This algorithm starts from an initial tree built by a fast distancebased method and modifies this tree to improve its likelihood at each iteration. Due to this simultaneous adjustment of the topology and branch lengths, only a few iterations are sufficient to reach an optimum. We used extensive and realistic computer simulations to show that the topological accuracy of this new method is at least as high as that of the existing maximumlikelihood programs and much higher than the performance of distancebased and parsimony approaches. The reduction of computing time is dramatic in comparison with other maximumlikelihood packages, while the likelihood maximization ability tends to be higher. For example, only 12 min were required on a standard personal computer to analyze a data set consisting of 500 rbcL sequences with 1,428 base pairs from plant plastids, thus reaching a speed of the same order as some popular distancebased and parsimony algorithms. This new method is implemented in the PHYML program, which is freely available on our web page:
Efficient Tree Searches with Available Algorithms
, 2007
"... Phylogenetic methods based on optimality criteria are highly desirable for their logic properties, but timeconsuming when compared to other methods of tree construction. Traditionally, researchers have been limited to exploring tree space by using multiple replicates of Wagner addition followed by ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
Phylogenetic methods based on optimality criteria are highly desirable for their logic properties, but timeconsuming when compared to other methods of tree construction. Traditionally, researchers have been limited to exploring tree space by using multiple replicates of Wagner addition followed by typical hill climbing algorithms such as SPR or/and TBR branch swapping but these methods have been shown to be insuficient for “large” data sets (or even for small data sets with a complex tree space). Here, I review different algorithms and search strategies used for phylogenetic analysis with the aim of clarifying certain aspects of this important part of the phylogenetic inference exercise. The techniques discussed here apply to both major families of methods based on optimality criteria—parsimony and maximum likelihood—and allow the thorough analysis of complex data sets with hundreds to thousands of terminal taxa. A new technique, called preprocessed searches is proposed for reusing phylogenetic results obtained in previous analyses, to increase the applicability of the previously proposed jumpstarting phylogenetics method. This article is aimed to serve as an educational and algorithmic reference to biologists interested in phylogenetic analysis. Rationale In phylogenetic analysis, numerical methods are preferred over other methods because of their efficiency and repeatability. Within numerical methods, those based on optimality criteria are to be preferred because they allow for hypothesis testing and tree comparisons based on objective measures. However,
A GLOBAL MAXIMUM LIKELIHOOD SUPERQUARTET PHYLOGENY METHOD
"... Extending the idea of our previous algorithm [17, 18] we developed a new sequential quartetbased phylogenetic tree construction method. This new algorithm reconstructs the phylogenetic tree iteratively by examining at each merge step every possible superquartet which is formed by four subtrees ins ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Extending the idea of our previous algorithm [17, 18] we developed a new sequential quartetbased phylogenetic tree construction method. This new algorithm reconstructs the phylogenetic tree iteratively by examining at each merge step every possible superquartet which is formed by four subtrees instead of simple quartet in our previous algorithm. Because our new algorithm evaluates superquartet trees, each of which may consist of more than four molecular sequences, it can effectively alleviate a traditional, but important problem of quartet errors encountered in the quartetbased methods. Experiment results show that our newly proposed algorithm is capable of achieving very high accuracy and solid consistency in reconstructing the phylogenetic trees on different sets of synthetic DNA data under various evolution circumstances. 1
Informationtheoretic indices and an approximate significance test for testing the molecular clock hypothesis with genetic distances. Mol Phylogenet Evol
"... a b s t r a c t Distancebased phylogenetic methods are widely used in biomedical research. However, distancebased dating of speciation events and the test of the molecular clock hypothesis are relatively underdeveloped. Here I develop an approximate test of the molecular clock hypothesis for dist ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
a b s t r a c t Distancebased phylogenetic methods are widely used in biomedical research. However, distancebased dating of speciation events and the test of the molecular clock hypothesis are relatively underdeveloped. Here I develop an approximate test of the molecular clock hypothesis for distancebased trees, as well as informationtheoretic indices that have been used frequently in model selection, for use with distance matrices. The results are in good agreement with the conventional sequencebased likelihood ratio test. Among the informationtheoretic indices, AICu is the most consistent with the sequencebased likelihood ratio test. The confidence in model selection by the indices can be evaluated by bootstrapping. I illustrate the usage of the indices and the approximate significance test with both empirical and simulated sequences. The tests show that distance matrices from protein gel electrophoresis and from genome rearrangement events do not violate the molecular clock hypothesis, and that the evolution of the third codon position conforms to the molecular clock hypothesis better than the second codon position in vertebrate mitochondrial genes. I outlined evolutionary distances that are appropriate for phylogenetic reconstruction and dating.
PowerPoint®
"... image downloads Journal information Published on behalf of The full text of this article, along with updated information and services is available online at ..."
Abstract
 Add to MetaCart
(Show Context)
image downloads Journal information Published on behalf of The full text of this article, along with updated information and services is available online at
DOI: 10.1080/10635150390196993 Accelerated Likelihood Surface Exploration: The Likelihood Ratchet
"... Abstract. — The existence of multiple likelihood maxima necessitates algorithms that explore a large part of the tree space. However, because of computational constraints, stepwise additionbased treesearching methods do not allow for this exploration in reasonable time. Here, I present an algorith ..."
Abstract
 Add to MetaCart
Abstract. — The existence of multiple likelihood maxima necessitates algorithms that explore a large part of the tree space. However, because of computational constraints, stepwise additionbased treesearching methods do not allow for this exploration in reasonable time. Here, I present an algorithm that increases the speed at which the likelihood landscape can be explored. The iterative algorithm combines the computational speed of distancebased tree construction methods to arrive at approximations of the global optimum with the accuracy of optimality criterion based branchswapping methods to improve on the result of the starting tree. The algorithm moves between local optima by iteratively perturbing the tree landscape through a process of reweighting randomly drawn samples of the underlying sequence data set. Tests on simulated and real data sets demonstrated that the optimal solution obtained using stepwise additionbased heuristic searches was found faster using the algorithm presented here. Tests on a previously published data set that established the presence of tree islands under maximum likelihood demonstrated that the algorithm identifies the same tree islands in a shorter amount of time than that needed using stepwise addition. The algorithm can be readily applied using standard software for phylogenetic inference. [Heuristics; maximum likelihood; parsimony ratchet; phylogenetic inference; tree landscapes.] Algorithms for finding maximumlikelihood trees can sometimes find multiple solutions (Steel, 1994; Rogers and Swofford, 1999; Chor et al., 2000; Salter, 2001), necessitating
NJML+P: A Hybrid Algorithm of the Maximum Likelihood and NeighborJoining Methods Using Parallel Computing
"... ..."
(Show Context)
NJML+P: A Hybrid Algorithm of the Maximum Likelihood and NeighborJoining Methods Using Parallel Computing
"... The NJML method [2, 3] is a hybrid algorithm of the two wellknown methods to reconstruct molecular phylogenetic trees: the neighborjoining (NJ) method [4] and the maximum likelihood (ML) method [1]. The NJML method is considerably efficient both in reliability and speed comparing with the other ex ..."
Abstract
 Add to MetaCart
(Show Context)
The NJML method [2, 3] is a hybrid algorithm of the two wellknown methods to reconstruct molecular phylogenetic trees: the neighborjoining (NJ) method [4] and the maximum likelihood (ML) method [1]. The NJML method is considerably efficient both in reliability and speed comparing with the other existing MLbased methods. By giving appropriate parameters, the NJML method gradually