#### DMCA

## Evolutionary inference via the Poisson indel process (2012)

Venue: | Proc. Natl. Acad. Sci., 10.1073/pnas.1220450110 |

Citations: | 7 - 2 self |

### Citations

6493 |
The neighbor-joining method: a new method for reconstructing phylogenetic trees
- Saitou, Nei
(Show Context)
Citation Context ...ical understanding, the difficulty of getting calibrated confidence intervals, and over-alignment problems [56, 35]. Finally, other methods have focused on analyzing only pairs of sequences at a time =-=[53, 51, 56, 6]-=-. While this approach can considerably simplify computation [20, 9], it has the disadvantage that it is not based on an underlying joint posterior probability distribution. In the current paper we pre... |

2519 |
A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences
- Kimura
- 1980
(Show Context)
Citation Context ....33 tion, branch lengths sampled from rate 2 exponential distributions, indels generated from the PIP with parameters η = 100, ζ = 1, and nucleotides sampled from the Kimura two-parameter model (K2P) =-=[26]-=-. We measured the quality of MSA reconstructions using the F1 score, defined as the harmonic mean of the reconstructed alignment edge recall (called the sum-of-pairs score or developer’s score in the ... |

2179 | A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst
- Guindon, Gascuel
- 2003
(Show Context)
Citation Context ...d tree. We used synthetic data to assess the quality of the tree reconstructions produced by PIP, compared to the reconstructions of PhyML 2.4.4, a widelyused platform for phylogenetic tree inference =-=[15]-=-. We also compared the inferred MSAs to those produced by Clustal 2.0.12 [19], a popular MSA inference system. While our implementation evaluated in this section is based on the Bayesian framework, we... |

1883 |
Evolutionary trees from DNA sequences: a maximum likelihood approach
- Felsenstein
(Show Context)
Citation Context ...erally based on the assumption that the columns of the alignment are independent, in which case the problem decouples into a simple recursion on the tree (the “Felsenstein” or “sum-product” recursion =-=[13]-=-). Such an approach can introduce numerous artifacts, however, both in the inferred phylogeny [46, 47, 67], and in the inferred alignment [52, 43]. It is also possible to iterate the solution of the M... |

1478 |
Exact stochastic simulation of coupled chemical reactions.
- Gillespie
- 1977
(Show Context)
Citation Context ...ction 6. 2 Background We begin by giving a brief overview of the TKF91 model. Instead of following the standard treatment based on differential equations, we present a DoobGillespie view of the model =-=[11, 14]-=- that will be useful in our subsequent development. Let us assume that at some point in time t, a sequence has length n. In the TKF91 process, the sequence stays unchanged for a random interval of tim... |

655 | Markov Models in computational biology : applications to protein modeling
- Krogh, Brown, et al.
- 1994
(Show Context)
Citation Context ...onal marginals are therefore possible [21, 17, 61, 37, 18, 40, 36, 45]. This has the appeal that statistical inference under these processes (known as transducers) can be based on dynamic programming =-=[2, 29, 57, 42]-=-. Unfortunately, however, despite some analytic simplification that is feasible in restricted cases [60], the memory needed to represent the state space in these models is generally exponential in the... |

584 |
CLUSTAL: a package for performing multiple sequence alignment on a microcomputer. Gene 73:237–244
- DG, PM
- 1988
(Show Context)
Citation Context ...ons produced by PIP, compared to the reconstructions of PhyML 2.4.4, a widelyused platform for phylogenetic tree inference [15]. We also compared the inferred MSAs to those produced by Clustal 2.0.12 =-=[19]-=-, a popular MSA inference system. While our implementation evaluated in this section is based on the Bayesian framework, we evaluate it using a frequentist methodology. More precisely, we use Bayes es... |

170 |
An evolutionary model for maximum likelihood alignment of DNA sequences.
- JL, Kishino, et al.
- 1991
(Show Context)
Citation Context ...olecular sequences. This problem is generally formulated in terms of stringvalued evolutionary processes along the branches of a phylogenetic tree. The classical evolutionary process, the TKF91 model =-=[62]-=-, is a continuous-time Markov chain model comprised of insertion, deletion and substitution events. Unfortunately this model gives rise to an intractable computational problem—the computation of the m... |

160 | The Bayesian Choice: from decision-theoretic foundations to computational implementation - Robert - 2001 |

156 | Minimal mutation trees of sequences
- Sankoff
- 1975
(Show Context)
Citation Context ...led effects on the overall inference. A second approach is to consider joint models that are not obtained by marginalization of a joint continuous-time string-valued process. A range of combinatorial =-=[54, 66, 31, 64, 58, 34]-=- and probabilistic [16, 28, 48, 46, 47] models fall in this category. Although often inspired by continuous-time processes, obtaining a coherent and calibrated estimate of uncertainty in these models ... |

119 |
Inching toward reality: an improved likelihood model of sequence evolution.
- JL, Kishino, et al.
- 1992
(Show Context)
Citation Context ...ial power of the basic models. For example, there has been significant work on extending TKF91 to models that capture the “long indels” that arise biologically but are not captured by the basic model =-=[63, 41, 40]-=-. In this regard, we wish to note that the Poisson representation of the PIP model provides new avenues for extension that are not available within the TKF91 framework. In particular, the superpositio... |

95 |
D: A Compound Poisson Process for Relaxing the Molecular Clock
- JP, Larget, et al.
(Show Context)
Citation Context ...ral states) can be computed in time linear in the number of taxa, rather than exponential as in the case of TKF91. Poisson representations have played an important role in pure substitution processes =-=[22, 44, 1]-=-, but to the best of knowledge, this is the first time Poisson representations are used for indel inference. Although the insertion process in TKF91 might be argued to be more realistic biologically t... |

92 |
Some statistical methods connected with series of events
- Cox
- 1955
(Show Context)
Citation Context ... to improve PIP models is to make the insertion rate mean measure more realistic: instead of being uniform across the tree, it could be modeled using a prior distribution, hence forming a Cox process =-=[8]-=-. This would be most useful when the sequences under study have large length or indel intensity variations across sites and branches [59]. Acknowledgments We would like to thank Bastien Boussau, Ian H... |

77 |
Evolutionary HMMs: a Bayesian approach to multiple alignment
- Holmes, Bruno
- 2001
(Show Context)
Citation Context ...KF91 model can be represented as a hidden Markov model (HMM), and that generalizations to a broader class of string-valued stochastic processes with finitedimensional marginals are therefore possible =-=[21, 17, 61, 37, 18, 40, 36, 45]-=-. This has the appeal that statistical inference under these processes (known as transducers) can be based on dynamic programming [2, 29, 57, 42]. Unfortunately, however, despite some analytic simplif... |

66 |
Bayesian coestimation of phylogeny and sequence alignment
- Lunter, Miklos, et al.
- 2005
(Show Context)
Citation Context ...KF91 model can be represented as a hidden Markov model (HMM), and that generalizations to a broader class of string-valued stochastic processes with finitedimensional marginals are therefore possible =-=[21, 17, 61, 37, 18, 40, 36, 45]-=-. This has the appeal that statistical inference under these processes (known as transducers) can be based on dynamic programming [2, 29, 57, 42]. Unfortunately, however, despite some analytic simplif... |

64 | Large-Scale comparison of Protein Sequence alignment Algorithms
- Sauder, Arthur, et al.
(Show Context)
Citation Context ...he quality of MSA reconstructions using the F1 score, defined as the harmonic mean of the reconstructed alignment edge recall (called the sum-of-pairs score or developer’s score in the MSA literature =-=[55]-=-) and alignment edge precision (modeler’s score [68]). We measured the quality of tree reconstructions using the partition (symmetric clade difference) metric [5] and the weighted Robinson-Foulds metr... |

62 |
A ”long indel” model for evolutionary sequence alignment
- Miklos, Lunter, et al.
(Show Context)
Citation Context ...KF91 model can be represented as a hidden Markov model (HMM), and that generalizations to a broader class of string-valued stochastic processes with finitedimensional marginals are therefore possible =-=[21, 17, 61, 37, 18, 40, 36, 45]-=-. This has the appeal that statistical inference under these processes (known as transducers) can be based on dynamic programming [2, 29, 57, 42]. Unfortunately, however, despite some analytic simplif... |

59 |
Joint Bayesian estimation of alignment and phylogeny
- Redelings, Suchard
- 2005
(Show Context)
Citation Context ...ond approach is to consider joint models that are not obtained by marginalization of a joint continuous-time string-valued process. A range of combinatorial [54, 66, 31, 64, 58, 34] and probabilistic =-=[16, 28, 48, 46, 47]-=- models fall in this category. Although often inspired by continuous-time processes, obtaining a coherent and calibrated estimate of uncertainty in these models is difficult. A third possible response... |

48 |
Alignment Uncertainty and Genomic Analysis,
- Wong, Suchard, et al.
- 2008
(Show Context)
Citation Context ...e problem decouples into a simple recursion on the tree (the “Felsenstein” or “sum-product” recursion [13]). Such an approach can introduce numerous artifacts, however, both in the inferred phylogeny =-=[46, 47, 67]-=-, and in the inferred alignment [52, 43]. It is also possible to iterate the solution of the MSA problem and the tree inference problem [32, 33], which can be viewed as a heuristic methodology for att... |

42 |
Comparison of weighted labelled trees
- Robinson, Foulds
- 1979
(Show Context)
Citation Context ...nd alignment edge precision (modeler’s score [68]). We measured the quality of tree reconstructions using the partition (symmetric clade difference) metric [5] and the weighted Robinson-Foulds metric =-=[50]-=-. Relative improvements were obtained by computing the absolute value of the quality difference (in terms of the F1 for alignments, and Robinson-Foulds distance for trees), divided by the initial valu... |

40 |
Rapid and accurate large-scale coestimation of sequence alignments and phylogenetic trees
- Liu, Raghavan, et al.
- 2009
(Show Context)
Citation Context ... artifacts, however, both in the inferred phylogeny [46, 47, 67], and in the inferred alignment [52, 43]. It is also possible to iterate the solution of the MSA problem and the tree inference problem =-=[32, 33]-=-, which can be viewed as a heuristic methodology for attempting to perform joint inference. The drawbacks of these systems include a lack of theoretical understanding, the difficulty of getting calibr... |

39 |
Finite-state models in the alignment of macromolecules
- Allison
- 1992
(Show Context)
Citation Context ...onal marginals are therefore possible [21, 17, 61, 37, 18, 40, 36, 45]. This has the appeal that statistical inference under these processes (known as transducers) can be based on dynamic programming =-=[2, 29, 57, 42]-=-. Unfortunately, however, despite some analytic simplification that is feasible in restricted cases [60], the memory needed to represent the state space in these models is generally exponential in the... |

39 |
The evolution of non-coding chloroplast DNA and its applica‐ tion in plant systematics.
- Kelchner
- 2000
(Show Context)
Citation Context ...ion property of Poisson 16 processes makes it possible to combine the PIP model with other models that follow a Poisson law. For example, if the location X ′ of long indels, slipped-strand mispairing =-=[25]-=- or other non-local changes follow a Poisson point process, the union U = X ∪X ′ of the non-local changes with the point indels X provided by a PIP will also be distributed according to a Poisson proc... |

38 |
A probabilistic model for the evolution of RNA structure
- Holmes
- 2004
(Show Context)
Citation Context ...ls, and over-alignment problems [56, 35]. Finally, other methods have focused on analyzing only pairs of sequences at a time [53, 51, 56, 6]. While this approach can considerably simplify computation =-=[20, 9]-=-, it has the disadvantage that it is not based on an underlying joint posterior probability distribution. In the current paper we present a new approach to the joint probabilistic inference of trees a... |

36 |
Fast statistical alignment
- Bradley, Roberts, et al.
- 2009
(Show Context)
Citation Context ...ical understanding, the difficulty of getting calibrated confidence intervals, and over-alignment problems [56, 35]. Finally, other methods have focused on analyzing only pairs of sequences at a time =-=[53, 51, 56, 6]-=-. While this approach can considerably simplify computation [20, 9], it has the disadvantage that it is not based on an underlying joint posterior probability distribution. In the current paper we pre... |

36 |
POY version 4: phylogenetic analysis using dynamic homologies.
- VARON, VINH, et al.
- 2009
(Show Context)
Citation Context ...led effects on the overall inference. A second approach is to consider joint models that are not obtained by marginalization of a joint continuous-time string-valued process. A range of combinatorial =-=[54, 66, 31, 64, 58, 34]-=- and probabilistic [16, 28, 48, 46, 47] models fall in this category. Although often inspired by continuous-time processes, obtaining a coherent and calibrated estimate of uncertainty in these models ... |

36 |
Protein-length distributions for the three domains of life.
- Zhang
- 2000
(Show Context)
Citation Context ...bution represents a modeling advantage of PIP over TKF91, which has a geometrically distributed stationary distribution. Based on a study of protein-length distributions for the three domains of life =-=[69]-=-, the Poisson distribution has been suggested [38] as a more adequate length distribution. From Proposition 2, we can also obtain an alternative reparameterization of the PIP model, in terms of asympt... |

35 |
Evolutionary models for insertions and deletions in a probabilistic modeling framework
- Rivas
- 2005
(Show Context)
Citation Context ...ond approach is to consider joint models that are not obtained by marginalization of a joint continuous-time string-valued process. A range of combinatorial [54, 66, 31, 64, 58, 34] and probabilistic =-=[16, 28, 48, 46, 47]-=- models fall in this category. Although often inspired by continuous-time processes, obtaining a coherent and calibrated estimate of uncertainty in these models is difficult. A third possible response... |

30 |
Multiple alignment by sequence annealing
- Schwartz, Pachter
- 2007
(Show Context)
Citation Context ... dynamic program. Indeed, the running time of the most sophisticated algorithm for computing marginals [39] depends on the number of homology linearizations, which is exponential in sparse alignments =-=[56]-=-. As a consequence of this unfavorable computational complexity, there has been extensive work on approximations, specifically on approximations to the joint marginal probability of a tree and an alig... |

28 | Latent-variable modeling of string transductions with finite-state methods,” in
- Dreyer, Smith, et al.
- 2008
(Show Context)
Citation Context ...ite some analytic simplification that is feasible in restricted cases [60], the memory needed to represent the state space in these models is generally exponential in the number of leaves in the tree =-=[12]-=-. Moreover, even in the simple TKF91 model, there does not appear to be additional structure in the state space that allows for simplification of the 2 AC Gv3 :AT T A Cv2 : AC T Cv3 : TG C Cv1 : TA CΩ... |

27 |
Sequence alignments and pair hidden markov models using evolutionary history
- Knudsen, Miyamoto
(Show Context)
Citation Context ...ond approach is to consider joint models that are not obtained by marginalization of a joint continuous-time string-valued process. A range of combinatorial [54, 66, 31, 64, 58, 34] and probabilistic =-=[16, 28, 48, 46, 47]-=- models fall in this category. Although often inspired by continuous-time processes, obtaining a coherent and calibrated estimate of uncertainty in these models is difficult. A third possible response... |

26 | StatAlign: an extendable software package for joint Bayesian estimation of alignments and evolutionary trees
- Novak, Miklos, et al.
- 2008
(Show Context)
Citation Context |

26 |
MALIGN: A multiple sequence alignment program.
- Wheeler, Gladstein
- 1994
(Show Context)
Citation Context ...led effects on the overall inference. A second approach is to consider joint models that are not obtained by marginalization of a joint continuous-time string-valued process. A range of combinatorial =-=[54, 66, 31, 64, 58, 34]-=- and probabilistic [16, 28, 48, 46, 47] models fall in this category. Although often inspired by continuous-time processes, obtaining a coherent and calibrated estimate of uncertainty in these models ... |

25 | An algorithm for statistical alignment of sequences related by a binary tree.
- Hein
- 2002
(Show Context)
Citation Context |

23 | Automata-theoretic models of mutation and alignment.
- Searls, Murphy
- 1995
(Show Context)
Citation Context ...onal marginals are therefore possible [21, 17, 61, 37, 18, 40, 36, 45]. This has the appeal that statistical inference under these processes (known as transducers) can be based on dynamic programming =-=[2, 29, 57, 42]-=-. Unfortunately, however, despite some analytic simplification that is feasible in restricted cases [60], the memory needed to represent the state space in these models is generally exponential in the... |

22 |
SATe-II: very fast and accurate simultaneous estimation of multiple sequence alignments and phylogenetic trees
- Liu, Warnow, et al.
- 2012
(Show Context)
Citation Context ... artifacts, however, both in the inferred phylogeny [46, 47, 67], and in the inferred alignment [52, 43]. It is also possible to iterate the solution of the MSA problem and the tree inference problem =-=[32, 33]-=-, which can be viewed as a heuristic methodology for attempting to perform joint inference. The drawbacks of these systems include a lack of theoretical understanding, the difficulty of getting calibr... |

21 | Applying the Thorne-Kishino-Felsenstein model to sequence evolution on a star-shaped tree
- Steel, Hein
- 2001
(Show Context)
Citation Context |

20 |
Recursions for statistical multiple alignment.
- Hein, Jensen, et al.
- 2003
(Show Context)
Citation Context |

19 | A generalized affine gap model significantly improves protein sequence alignment accuracy
- Zachariah, Crooks, et al.
- 2005
(Show Context)
Citation Context ..., defined as the harmonic mean of the reconstructed alignment edge recall (called the sum-of-pairs score or developer’s score in the MSA literature [55]) and alignment edge precision (modeler’s score =-=[68]-=-). We measured the quality of tree reconstructions using the partition (symmetric clade difference) metric [5] and the weighted Robinson-Foulds metric [50]. Relative improvements were obtained by comp... |

16 |
Toward extracting all phylogenetic information from matrices of evolutionary distances
- Roch
(Show Context)
Citation Context ...ical understanding, the difficulty of getting calibrated confidence intervals, and over-alignment problems [56, 35]. Finally, other methods have focused on analyzing only pairs of sequences at a time =-=[53, 51, 56, 6]-=-. While this approach can considerably simplify computation [20, 9], it has the disadvantage that it is not based on an underlying joint posterior probability distribution. In the current paper we pre... |

13 | Phylogenetic inference via Sequential Monte Carlo. Systematic Biology,
- Bouchard-Cote, Sankararaman, et al.
- 2011
(Show Context)
Citation Context ...ated MCMC kernels, some of which could be applied to inference in our model, for example [46]. Another potential direction would be to replace MCMC by a Sequential Monte Carlo posterior approximation =-=[4]-=-. It should also be emphasized that point indels is certainly not the exclusive driving force behind sequence evolution. In particular, “long indels” (atomic insertions and deletions of long segments,... |

13 | Statistical alignment: Recent progress, new applications, and challenges
- Lunter, Drummond, et al.
- 2004
(Show Context)
Citation Context ...tempting to perform joint inference. The drawbacks of these systems include a lack of theoretical understanding, the difficulty of getting calibrated confidence intervals, and over-alignment problems =-=[56, 35]-=-. Finally, other methods have focused on analyzing only pairs of sequences at a time [53, 51, 56, 6]. While this approach can considerably simplify computation [20, 9], it has the disadvantage that it... |

12 | Gibbs sampler for statistical multiple alignment - Jensen, Hein - 2002 |

12 | Accurate extension of multiple sequence alignments using a phylogeny-aware graph algorithm. Bioinformatics 28:1684–1691
- Löytynoja, AJ, et al.
- 2012
(Show Context)
Citation Context |

10 | Phylogenetic profiling of insertions and deletions in vertebrate genomes.
- Snir, Pachter
- 2006
(Show Context)
Citation Context ...modeled using a prior distribution, hence forming a Cox process [8]. This would be most useful when the sequences under study have large length or indel intensity variations across sites and branches =-=[59]-=-. Acknowledgments We would like to thank Bastien Boussau, Ian Holmes, Michael Newton and Marc Suchard for their comments and suggestions. This work was partially supported by a grant from the Office o... |

9 |
Markoff chains – denumerable case.
- Doob
- 1945
(Show Context)
Citation Context ...ction 6. 2 Background We begin by giving a brief overview of the TKF91 model. Instead of following the standard treatment based on differential equations, we present a DoobGillespie view of the model =-=[11, 14]-=- that will be useful in our subsequent development. Let us assume that at some point in time t, a sequence has length n. In the TKF91 process, the sequence stays unchanged for a random interval of tim... |

9 | Effect of non-independent substitution on phylogenetic accuracy. Syst. Biol
- Huelsenbeck, Nielsen
- 1999
(Show Context)
Citation Context ...uence of real numbers 0 < · · · < r′ < · · · < 1 is maintained. 7 In the first step, depicted in Figure 3A, a multiset of insertion points is sampled from a Poisson process defined on the phylogeny τ =-=[23]-=-. The rate measure for this Poisson process has atomic mass at the root of the tree; hence the need for multisets rather than simple point sets. Except for the root, no other points on the tree have a... |

9 |
An improved model for statistical alignment
- Miklós, Toroczkai
- 2001
(Show Context)
Citation Context ...ial power of the basic models. For example, there has been significant work on extending TKF91 to models that capture the “long indels” that arise biologically but are not captured by the basic model =-=[63, 41, 40]-=-. In this regard, we wish to note that the Poisson representation of the PIP model provides new avenues for extension that are not available within the TKF91 framework. In particular, the superpositio... |

9 | The effect of the guide tree on multiple sequence alignments and subsequent phylogenetic analyses. Pac Symp Biocomput
- Nelesen, Liu, et al.
- 2008
(Show Context)
Citation Context ...on the tree (the “Felsenstein” or “sum-product” recursion [13]). Such an approach can introduce numerous artifacts, however, both in the inferred phylogeny [46, 47, 67], and in the inferred alignment =-=[52, 43]-=-. It is also possible to iterate the solution of the MSA problem and the tree inference problem [32, 33], which can be viewed as a heuristic methodology for attempting to perform joint inference. The ... |

8 | Wagner and Dollo: a stochastic duet by composing two parsimonious solos. Syst. Biol
- Alekseyenko, Lee, et al.
- 2008
(Show Context)
Citation Context ...slipped-strand mispairing phenomena. Under the Poisson process representation, another interesting perspec4 tive on our process is to view it as a string-valued counterpart to stochastic Dollo models =-=[1, 44]-=-, which are defined on finite state spaces. In particular, the general idea of the two steps generation process used in Section 3 has antecedents in the literature on probabilistic modeling of morphol... |

8 | Efficient inference in phylogenetic indel trees - Bouchard-Côté, Jordan, et al. - 2008 |

8 |
Arbres de Steiner et réseaux dont certains sommets sont à localisation variable
- Bourque
- 1978
(Show Context)
Citation Context ...eloper’s score in the MSA literature [55]) and alignment edge precision (modeler’s score [68]). We measured the quality of tree reconstructions using the partition (symmetric clade difference) metric =-=[5]-=- and the weighted Robinson-Foulds metric [50]. Relative improvements were obtained by computing the absolute value of the quality difference (in terms of the F1 for alignments, and Robinson-Foulds dis... |

8 | R: GESTALT: Genomic steiner alignments
- Lancia, Ravi
- 1999
(Show Context)
Citation Context |

8 |
Incorporating Indel Information into Phylogeny Estimation for Rapidly Emerging Pathogens.
- Redelings, Suchard
- 2007
(Show Context)
Citation Context |

7 |
Transition probabilities for general birth-death processes with applications in ecology, genetics, and evolution
- Crawford, Suchard
(Show Context)
Citation Context ...ls, and over-alignment problems [56, 35]. Finally, other methods have focused on analyzing only pairs of sequences at a time [53, 51, 56, 6]. While this approach can considerably simplify computation =-=[20, 9]-=-, it has the disadvantage that it is not based on an underlying joint posterior probability distribution. In the current paper we present a new approach to the joint probabilistic inference of trees a... |

7 | 2012 accurate reconstruction of insertion-deletion histories by statistical phylogenetics
- Westesson, Lunter, et al.
(Show Context)
Citation Context ...l complexity, there has been extensive work on approximations, specifically on approximations to the joint marginal probability of a tree and an alignment, obtained by integrating over the derivation =-=[36, 65]-=-. A difficulty, however, is that these marginal probabilities play a role in tree inference procedures as the numerators and denominators of acceptance probabilities for Markov chain Monte Carlo algor... |

6 |
Handbook of Weighted Automata, chapter 6
- Mohri
- 2009
(Show Context)
Citation Context |

5 |
A unified approach to phylogenies and alignments
- Hein
- 1990
(Show Context)
Citation Context |

5 |
Improving progressive alignment for phylogeny reconstruction using parsimonious guide-trees
- Roshan, Livesay, et al.
(Show Context)
Citation Context ...on the tree (the “Felsenstein” or “sum-product” recursion [13]). Such an approach can introduce numerous artifacts, however, both in the inferred phylogeny [46, 47, 67], and in the inferred alignment =-=[52, 43]-=-. It is also possible to iterate the solution of the MSA problem and the tree inference problem [32, 33], which can be viewed as a heuristic methodology for attempting to perform joint inference. The ... |

3 |
Quasi-stationary distributions on continuous Markov chains
- Buiculescu
- 1972
(Show Context)
Citation Context ...e set, for all σ, σ′ ∈ Σε: ν( dx) = λ ( τ( dx) + 1 µ δΩ( dx) ) , Qσ,σ′ = −∑σ′′ 6=σ′ Qσ,σ′′ if σ = σ′ 0 if σ = ε µ if σ′ = ε θσ,σ′ o.w., and set pi to be the quasi-stationary distribution of Q =-=[7]-=-. The proof is given in the Appendix A. Note that in the case of interest here, where the rate of deletion does not depend on the character being deleted, piσ is equal to the entry of the stationary d... |

3 |
von Haeseler A. Assessing variability by joint sampling of alignments and mutation rates
- Metzler, Fleissner, et al.
(Show Context)
Citation Context |

3 |
Algorithms in Bioinformatics, chapter Bayesian Phylogenetic Inference under a Statistical Insertion-Deletion Model
- Mikls, Drummond, et al.
- 2003
(Show Context)
Citation Context ... substitution events. Unfortunately this model gives rise to an intractable computational problem—the computation of the marginal likelihood under the TKF91 model is exponential in the number of taxa =-=[39]-=-. In this work, we present a new stochastic process, the Poisson Indel Process (PIP), in which the complexity of this computation is reduced to linear. The new model is closely related to the TKF91 mo... |

2 |
Algorithm for statistical alignment of sequences derived from a Poisson sequence length distribution
- Miklós
(Show Context)
Citation Context ... TKF91, which has a geometrically distributed stationary distribution. Based on a study of protein-length distributions for the three domains of life [69], the Poisson distribution has been suggested =-=[38]-=- as a more adequate length distribution. From Proposition 2, we can also obtain an alternative reparameterization of the PIP model, in terms of asymptotic expected length η = λ/µ and indel intensity ζ... |

2 | A sufficient condition for reducing recursions in hidden Markov models
- Song
(Show Context)
Citation Context ...nce under these processes (known as transducers) can be based on dynamic programming [2, 29, 57, 42]. Unfortunately, however, despite some analytic simplification that is feasible in restricted cases =-=[60]-=-, the memory needed to represent the state space in these models is generally exponential in the number of leaves in the tree [12]. Moreover, even in the simple TKF91 model, there does not appear to b... |

1 | Convergence of the simulated annealing algorithm - Delyon - 1988 |

1 |
Phylogenetic methods and the prehistory languages, chapter Quantifying uncertainty in a stochastic Dollo model of vocabulary evolution
- Nicholls, Gray
- 2006
(Show Context)
Citation Context ...slipped-strand mispairing phenomena. Under the Poisson process representation, another interesting perspec4 tive on our process is to view it as a string-valued counterpart to stochastic Dollo models =-=[1, 44]-=-, which are defined on finite state spaces. In particular, the general idea of the two steps generation process used in Section 3 has antecedents in the literature on probabilistic modeling of morphol... |

1 |
Tracing the most parsimonious indel history
- Snir, Pachter
(Show Context)
Citation Context |