mlphylo {ape}R Documentation

Estimating Phylogenies by Maximum Likelihood

Description

mlphylo estimates a phylogenetic tree by maximum likelihood given a set of DNA sequences. The model of evolution is specified with the function DNAmodel.

logLik, deviance, and AIC are generic functions used to extract the log-likelihood, the deviance (-2*log-likelihood), or the Akaike information criterion of a tree. If no such values are available, NULL is returned.

Usage

mlphylo(model = DNAmodel(), x, phy, search.tree = FALSE)
## S3 method for class 'phylo':
logLik(object, ...)
## S3 method for class 'phylo':
deviance(object, ...)
## S3 method for class 'phylo':
AIC(object, ..., k = 2)

Arguments

model an object of class "DNAmodel" giving the model to be fitted.
x a matrix, a data frame, or a list giving the (aligned) DNA sequence data.
phy an object of class "phylo" giving the (initial) tree.
search.tree a logical specifying whether to search for the best tree (defaults to FALSE) (not functional for the moment).
object an object of class "phylo".
k a numeric value giving the penalty per estimated parameter; the default is k = 2 which is the classical Akaike information criterion.
... further arguments passed to or from other methods.

Details

The present version is a pre-alpha release. All comments, suggestions, bug reports, are warmly welcome.

The model specified by DNAmodel is fitted using the standard ``pruning'' algorithm of Felsenstein (1981). An algorithm for the estimation of tree topology is under development, and will be released when ready.

The implementation of the inter-sites variation in substitution rates follows the methodology developed by Yang (1994).

The difference among partitions is parametrized with a contrast parameter (denoted xi) that specifies the contrast in mean susbtitution rate among the partitions. This methodology will be detailed in a forthcoming paper.

The substitution rates are indexed column-wise in the rate matrix: the first rate is set to one.

Value

an object of class "phylo" with branch lengths as estimated by the function. There are two additional attributes:

loglik the maximum log-likelihood.
para the estimated parameters for each partition.

Note

For the moment, it is not possible to estimate neither branch lengths, nor the topology with mlphylo: this is still in development and will be released when ready. The function may estimate all other parameters: substitution rates, shape (alpha) of the inter-sites variation in substitution rates, the proportion of invariants, and the ``contrast'' parameter (xi) among partitions.

Alternative topologies can also be compared using likelihood-ratio tests (LRTs) or AICs.

Author(s)

Emmanuel Paradis paradis@isem.univ-montp2.fr

References

Felsenstein, J. (1981) Evolutionary trees from DNA sequences: a maximum likelihood approach. Journal of Molecular Evolution, 17, 368–376.

Yang, Z. (1994) Maximum likelihood phylogenetic estimation from DNA sequences with variable rates over sites: approximate methods. Journal of Molecular Evolution, 39, 306–314.

See Also

DNAmodel, nj, read.dna, summary.phylo


[Package ape version 1.8-2 Index]