Question: Where has the R Ape 'chronogram' function gone (Non-Parametric Rate Smoothing)?
0
gravatar for BlastedBadger
4 months ago by
Thule
BlastedBadger50 wrote:

Hi,

I want to use several methods to estimate the age of nodes in a phylogeny. I plan on using Ape's chronos function that relies on a penalized likelihood method (Sanderson 2002), but I also would like to test the NPRS method (Non Parametric Rate Smoothing, Sanderson 1997). I read some tutorial where they say it's in the chronogram function, but I see no such function in Ape v4.1. Where is it? Was it removed?

ADD COMMENTlink modified 4 months ago • written 4 months ago by BlastedBadger50
2

This page may help: http://grokbase.com/t/r/r-sig-phylo/112as2y3m8/replacement-for-chronogram

ADD REPLYlink written 4 months ago by cpad01122.3k
1
gravatar for BlastedBadger
4 months ago by
Thule
BlastedBadger50 wrote:

From Ape's log, chronogram has indeed been deprecated.

CHANGES IN APE VERSION 2.5-2

DEPRECATED & DEFUNCT

o evolve.phylo() and plot.ancestral() have been removed.

o chronogram(), ratogram(), and NPRS.criterion() have been removed.

But as mentionned in this page (thanks @cpad0112), using the parameter lambda=0 in chronos is similar to applying the NPRS.

[EDIT] After looking at the function's code and confirming by reading Sanderson 2002, lambda is the level of penalization of the loglikelihood. So the higher the lambda, the closer you are from NPRS. I am pretty sure the page I quoted here is wrong. Here is the function of interest (penalized likelihood with correlated rate), in chronos:

correlated = function(rate,
        node.time) {
        loglik <- log.lik.poisson(rate, node.time)
        if (!is.finite(loglik)) return(-1e+100)
        loglik - lambda * (sum((rate[ind1] - rate[ind2])^2) +
            var(rate[basal]))
}
ADD COMMENTlink modified 3 months ago • written 4 months ago by BlastedBadger50
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 952 users visited in the last hour