## Performance study of phylogenetic methods: (unweighted) quartet methods and neighbor-joining (2003)

### Cached

### Download Links

- [www.cs.utexas.edu]
- [cs.unm.edu]
- [comet.lehman.cuny.edu]
- [www.cs.unm.edu]
- [www.cs.unm.edu]
- DBLP

### Other Repositories/Bibliography

Citations: | 35 - 5 self |

### BibTeX

@MISC{John03performancestudy,

author = {Katherine St. John and Tandy Warnow and Bernard M. E. Moret and Lisa Vawter},

title = { Performance study of phylogenetic methods: (unweighted) quartet methods and neighbor-joining},

year = {2003}

}

### OpenURL

### Abstract

### Citations

2544 |
The Neighbor-Joining method: a new method for reconstructing phylogenetic trees
- Saitou, Nei
- 1987
(Show Context)
Citation Context ...ods (methods that transform input sequences into a distance matrix and then construct the tree from that distance matrix), none is more widely used by biologists than the neighbor-joining (NJ) method =-=[22]-=-. Not only is it quite fast (O(n 3 ) for n taxa [24]), but experimental work has also shown that the trees NJ constructs are reasonably accurate, as long as the rate of evolution is neither too low no... |

1025 | Evolutionary trees from DNA sequences: a maximum likelihood approach - Felsenstein - 1981 |

871 | Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids - DURBIN, EDDY, et al. - 1998 |

860 | A simple method for estimating evolutionary rate of base substitutions through comparative analysis of nucleotide sequences - Kimura - 1980 |

602 | Biological Sequence Analysis - Durbin, Eddy, et al. - 1998 |

197 |
Seq-Gen: An application for the Monte Carlo simulation of DNA sequence evolution along phylogenetic trees
- RAMBAUT, GRASSLY
- 1997
(Show Context)
Citation Context ...m 10 7 to 10 3 . This process produces Jukes-Cantor trees with e values ranging from a minimum of 10 7 to a maximum of 1. We generated random DNA sequences for the root and used the program Seq-Gen [=-=20]-=- to evolve these sequences down the tree under the Jukes-Cantor model of evolution, thus producing sets of sequences at the leaves, our synthetic datasets. 4.3 Statistical considerations. Because the ... |

179 | Fast DNAml : a tool for construction of phylogenetics trees of DNA sequences using maximum likelihood - OLSEN, MATSUDA, et al. - 1994 |

102 | A few logs suffice to build (almost) all trees (part 2). Theoretical Computer Science - Erdos, Steel, et al. - 1999 |

84 |
The performance of the neighbor-joining methods of phylogenetic reconstruction. Algorithmica
- Atteson
- 1999
(Show Context)
Citation Context ...r tree reconstruction methods. 3 Theoretical Bounds on the Convergence Rates. We begin with the known upper bounds on the convergence rates of NJ and the Q method. Surprisingly, these are identical [=-=1, 6-=-], although experimental studies strongly suggest that NJ obtains accurate reconstructions of trees from shorter sequences than Q throughout the parameter space of Jukes-Cantor trees [11]. Theorem 3.... |

80 | Disk-Covering, A fast converging method for phylogenetic tree reconstruction
- Huson, Nettles, et al.
- 1999
(Show Context)
Citation Context ...om sequences that grow exponentially in n. While Theorem 3.1 provides only an upper bound, earlier experimental work shows that the Q method performs quite poorly when g and diam(T ) are both large [=-=12-=-], and that NJ is also aected, although less severely [11]. We now consider the convergence rate of the QC methods. Because the error bound used in QC methods is a multiple of p q e , the ratio of per... |

78 |
Success of phylogenetic methods in the four-taxon case
- Huelsenbeck, Hillis
- 1993
(Show Context)
Citation Context ...es of evolution, whereas NJ clearly outperforms ML for higher rates. Because our observations dier from the received wisdom in theseld, we oer the following possible explanation. In earlier studies [1=-=0]-=-, the performance of ML and NJ as quartet estimators was studied by explicitly simulating evolution on four-taxon trees. Here, we have simulated evolution on larger trees and then looked at the quarte... |

70 | Inferring evolutionary trees with strong combinatorial evidence
- Berry, Gascuel
- 1997
(Show Context)
Citation Context ...as the Buneman method) and the Quartet Cleaning methods, can be described in terms of an explicit bound on the number of quartet errors around the edges they reconstruct. We begin with the Q method [=-=3-=-]. This method seeks the maximally resolved tree T 0 obeying Q(T 0 ) Q. Therefore, there are no quartet errors around any edge in the tree T 0 . This tree always exists, since the star tree trivially... |

54 |
D: Distribution of tree comparison metricsâ€”Some new results. Syst. Biol
- MA, Penny
- 1993
(Show Context)
Citation Context ...rate on a random Jukes-Cantor tree. We require the following lemma. Lemma 3.1. The median diameter of all (2n 5)!! unrooted, leaf-labelled, binary trees on n leaves is ( p n). Proof. Penny and Steel [19] gave formulas for the distribution of interleaf distances in such trees under the assumption that all (2n 5)!! such trees are equally likely, obtaining (D) = 2 2n = 2n n = ( p n) and 2 (D) ... |

46 | A practical algorithm for recovering the best supported edges of an evolutionary tree (extended abstract
- Berry, Bryant, et al.
(Show Context)
Citation Context .... The Q tree is unique and can be constructed in polynomial time; by design, however, the Q method is conservative and generally produces very unresolved trees [11]. Quartet-Cleaning (QC) methods [2=-=, 4, 13]-=- have explicit bounds on the number of quartet errors around each reconstructed edge e. These error bounds have the form m p q e , where q e is the number of quartets around edge e and m is a small co... |

42 | Quartet cleaning: improved algorithms and simulations
- Berry, Jiang, et al.
- 1999
(Show Context)
Citation Context .... The Q tree is unique and can be constructed in polynomial time; by design, however, the Q method is conservative and generally produces very unresolved trees [11]. Quartet-Cleaning (QC) methods [2=-=, 4, 13]-=- have explicit bounds on the number of quartet errors around each reconstructed edge e. These error bounds have the form m p q e , where q e is the number of quartets around edge e and m is a small co... |

41 |
Approaches for assessing phylogenetic accuracy
- Hillis
- 1995
(Show Context)
Citation Context ...assessing the quality of phylogenetic reconstruction methods problematic (but see the study by Hillis et al. [8]). As a consequence, the method of choice for evaluating heuristics has been simulation =-=[7]. In -=-such a simulation, an ancestral biomolecular (DNA, RNA, or amino-acid) sequence is evolved along a \model" tree, producing a synthetic set of biomolecular sequences at the leaves. Phylogenetic re... |

38 |
Analyzing algorithms by simulation: variance reduction techniques and simulation speedups
- McGeoch
- 1992
(Show Context)
Citation Context ...er expanded by the choice of evolutionary rates, it is not possible to take a fair sample of the entire input space. In order to obtain statistically robust results, we followed the advice of McGeoch =-=[16]-=- and Moret [18] and used a number of runs, each composed of a number of trials (a trial is a single comparison), computed the mean outcome for each run, and studied the mean and standard deviation ove... |

37 | Towards a discipline of experimental algorithms
- Moret
- 1996
(Show Context)
Citation Context ...the choice of evolutionary rates, it is not possible to take a fair sample of the entire input space. In order to obtain statistically robust results, we followed the advice of McGeoch [16] and Moret =-=[18]-=- and used a number of runs, each composed of a number of trials (a trial is a single comparison), computed the mean outcome for each run, and studied the mean and standard deviation over the runs of t... |

27 | Quartet-based phylogenetic inference: Improvements and limits - Ranwez, Gascuel - 2001 |

20 |
A note on the neighborjoining method of saitou and nei. Molecular Biology and Evolution
- Studier, Keppler
- 1988
(Show Context)
Citation Context ...istance matrix and then construct the tree from that distance matrix), none is more widely used by biologists than the neighbor-joining (NJ) method [22]. Not only is it quite fast (O(n 3 ) for n taxa =-=[24]-=-), but experimental work has also shown that the trees NJ constructs are reasonably accurate, as long as the rate of evolution is neither too low nor too high. However, there is no comparative 2 SODA-... |

19 |
Evidence from betatubulin phylogeny that microsporidia evolved from within the fungi
- Keeling, Luker, et al.
- 2000
(Show Context)
Citation Context ...od and, in particular, whether the performance of QC methods scales with increasing problem size. Thesnal quartet-based method we examined is the best known and the most frequently used by biologists =-=[17, 21, 15]-=-: the Quartet-Puzzling (QP) method [23]. This heuristic computes quartets using ML and then uses a greedy strategy to construct a tree on which many input quartets are in agreement. QP uses an arbitra... |

17 |
Mammalian protein metabolism
- Jukes, Cantor
- 1969
(Show Context)
Citation Context .... C.), pp 186-195 study of NJ and quartet-based methods. We present the results of a detailed, large-scale experimental study of quartet-based methods and NJ under the Jukes-Cantor model of evolution =-=[14]-=-. Our results indicate that NJ always outperforms the quartet-based methods we examined, in terms of both accuracy and speed. We conclude that NJ, already the most popular distance-based method, shoul... |

15 | Hybrid tree reconstruction methods
- Huson, Nettles, et al.
- 1999
(Show Context)
Citation Context ...he constraint on any set of quartets. The Q tree is unique and can be constructed in polynomial time; by design, however, the Q method is conservative and generally produces very unresolved trees [1=-=1]-=-. Quartet-Cleaning (QC) methods [2, 4, 13] have explicit bounds on the number of quartet errors around each reconstructed edge e. These error bounds have the form m p q e , where q e is the number of ... |

14 | A polynomial-time approximation scheme for inferring evolutionary trees from quartet topologies and its application - Jiang, Kearney, et al. |

11 |
Quartet puzzling: a maximum likelihood method for reconstructing tree topologies
- Strimmer, Haeseler
- 1996
(Show Context)
Citation Context ...of QC methods scales with increasing problem size. Thesnal quartet-based method we examined is the best known and the most frequently used by biologists [17, 21, 15]: the Quartet-Puzzling (QP) method =-=[23]-=-. This heuristic computes quartets using ML and then uses a greedy strategy to construct a tree on which many input quartets are in agreement. QP uses an arbitrary ordering of taxa, constructs the opt... |

11 | A note on the neighbor-joining method of Saitou and - Studier, Kepler - 1988 |

8 |
Neuromedin U is a potent agonist at the orphan G protein-coupled receptor FM3
- Szekeres, Muir, et al.
- 2000
(Show Context)
Citation Context ...or research thrust in computational biology. The evolutionary process not only determines relationships among taxa, but also allows prediction of structural, physiological, and biochemical properties =-=[5, 25]-=-. Research on tree reconstruction has focused on reconstructing an evolutionary tree (phylogeny) under various optimization criteria. However, almost all optimization problems of interest to biologist... |

6 | A phylogeny of the damselfly genus Calopteryx (Odonata) using mitochondrial 16S rDNA markers. Molecular Phylogenetics and Evolution - Misof, Anderson, et al. - 2000 |

4 |
Molecular evolution and phylogeny of the buzzatii complex (Drosophila repleta group): a maximumlikelihood approach
- Rodriguez-Trelles, Alarcon, et al.
- 2000
(Show Context)
Citation Context ...od and, in particular, whether the performance of QC methods scales with increasing problem size. Thesnal quartet-based method we examined is the best known and the most frequently used by biologists =-=[17, 21, 15]-=-: the Quartet-Puzzling (QP) method [23]. This heuristic computes quartets using ML and then uses a greedy strategy to construct a tree on which many input quartets are in agreement. QP uses an arbitra... |

4 | von Haeseler A: Quartet puzzling: a maximum likelihood method for reconstructing tree topologies. Mol Biol Evol - Strimmer - 1996 |

2 |
A G proteincoupled receptor for Uridine 5â€™-diphosphoglucose (UDP-glucose
- Chambers, Macdonald, et al.
- 2000
(Show Context)
Citation Context ...or research thrust in computational biology. The evolutionary process not only determines relationships among taxa, but also allows prediction of structural, physiological, and biochemical properties =-=[5, 25]-=-. Research on tree reconstruction has focused on reconstructing an evolutionary tree (phylogeny) under various optimization criteria. However, almost all optimization problems of interest to biologist... |

2 |
A few logs suce to build almost all trees|II
- Erdos, Steel, et al.
- 1999
(Show Context)
Citation Context ...t to the tree T if it does not agree with T . If Q(T ) denotes the set of all quartets that agree with T , then T is uniquely characterized by Q(T ) and can be reconstructed from T in polynomial time =-=[6-=-]. Quartet-based methods operate in two phases. In thesrst phase they construct a set Q of quartets on the dierent sets of four taxa. One popular approach is to use the computationally intensive, but ... |

2 |
Experimental phylogenies: generation of a known phylogeny, Science 255
- Hillis, Bull, et al.
- 1992
(Show Context)
Citation Context ...? In biological applications, the true, historical tree is almost never known, which makes assessing the quality of phylogenetic reconstruction methods problematic (but see the study by Hillis et al. =-=[8]-=-). As a consequence, the method of choice for evaluating heuristics has been simulation [7]. In such a simulation, an ancestral biomolecular (DNA, RNA, or amino-acid) sequence is evolved along a \mode... |

2 |
A polynomial-time approximation scheme for inferring evolutionary trees from quartet topologies and its application
- unknown authors
(Show Context)
Citation Context ...ds have tosnd ways of handling incorrect quartets. Most optimization problems related to tree reconstruction from quartets are NP-hard. An example of this is the Maximum Quartet Compatibility problem =-=[13]-=-, which seeks a tree T for a given set Q of quartets such that jQ(T 0 ) \ Qj is maximized. The methods studied in this paper have no perforSODA-2001 (Washington, D. C.), pp 186-195 3 mance guarantees ... |

2 | An error-correcting map for quartest can improve the signals for phylogenetic trees - Willson - 2001 |

1 |
A phylogeny of the damsel genus Calopteryx (Odonata) using mitochondrial 16s rDNA markers
- Mishof, Anderson, et al.
(Show Context)
Citation Context ...od and, in particular, whether the performance of QC methods scales with increasing problem size. Thesnal quartet-based method we examined is the best known and the most frequently used by biologists =-=[17, 21, 15]-=-: the Quartet-Puzzling (QP) method [23]. This heuristic computes quartets using ML and then uses a greedy strategy to construct a tree on which many input quartets are in agreement. QP uses an arbitra... |