Auxiliary functions for providing optim.pml, pml_bb
fitting. Use it to construct a control or ratchet.par argument.
pml.control(epsilon = 1e-08, maxit = 10, trace = 1, tau = 1e-08,
statefreq = "empirical")ratchet.control(iter = 20L, maxit = 200L, minit = 100L, prop = 1/2,
rell = TRUE, bs = 1000L)
A list with components named as the arguments for controlling the fitting process.
Stop criterion for optimization (see details).
Maximum number of iterations (see details).
Show output during optimization (see details).
minimal edge length.
take "empirical" or "estimate" state frequencies.
Number of iterations to stop if there is no change.
Minimum number of iterations.
Only used if rearrangement=stochastic. How many NNI moves
should be added to the tree in proportion of the number of taxa.´
logical, if TRUE approximate bootstraping similar Minh et al. (2013) is performed.
number of approximate bootstrap samples.
Klaus Schliep klaus.schliep@gmail.com
pml.control controls the fitting process. epsilon and
maxit are only defined for the most outer loop, this affects
pmlCluster, pmlPart and pmlMix.
epsilon is not an absolute difference between, but instead is
defined as (logLik(k)-logLik(k+1))/logLik(k+1). This seems to be a good
compromise and to work reasonably well for small and large trees or
alignments.
If trace is set to zero than no out put is shown, if functions are
called internally than the trace is decreased by one, so a higher of trace
produces more feedback. It can be useful to figure out how long an run will
take and for debugging.
statefreq controls if base/state frequencies are optimized or
empirical estimates are taken, when this applies. For some nucleotide models
(e.g. JC, SYM) equal base frequencies and for amino acid models precomputed
state frequencies are used, if not '+F' is specified.
tau might be exactly zero if duplicated sequences in the alignment are
observed. In this case the analysis is performed only on unique sequences and
duplicated taxa are added to the tree with zero edge length. This may lead to
multifurcations if there are three or more identical sequences. After
optimization it is good practice to prune away edges of length tau
using di2multi. See also Janzen et al. (2021).
Minh, B. Q., Nguyen, M. A. T., & von Haeseler, A. (2013). Ultrafast approximation for phylogenetic bootstrap. Molecular biology and evolution, 30(5), 1188-1195.
Janzen, T., Bokma, F.,Etienne, R. S. (2021) Nucleotide Substitutions during Speciation may Explain Substitution Rate Variation, Systematic Biology, 71(5), 1244–1254.
pml_bb, optim.pml
pml.control()
pml.control(maxit=25)
Run the code above in your browser using DataLab