Hello, I'm trying to do ancestral reconstruction of a discrete character state using ace. I'm referencing the Ancestral State Reconstruction HowTo on the R phylo wiki (http://www.r-phylo.org/wiki/HowTo/Ancestral_State_Reconstruction). It says how to use a custom model with ace:
"What if you want to use a transition model that isn't hardwired into ace? You can do so by defining a custom transition matrix. For example, let's create a custom four-parameter matrix for a three state character, in which the the probability of transitioning from state 1 to 2 equals the probability of state 2 to 1 (parameter 1), the probability of transitioning from 1 to 3 equals the probability of transitioning from state 3 to 1 (parameter 2), but the probability of transitioning from state 2 to 3 (parameter 3) does not necessarily equal the probability of transitioning from state 3 to 2 (parameter 4)." Right below, on the page, it shows this matrix: [,1] [,2] [,3] [1,] 0 1 2 [2,] 1 0 4 [3,] 2 3 0 Isn't this the transpose of how model matrices normally work? I would expect "the probability of transitioning from state 2 to 3 (parameter 3)" to be in position [2,3], but it's in position [3,2] here. I realize that it doesn't make a difference in this case, but in my case I'm trying to define a non-reversible model so it does matter. So, sorry if this is a dumb question, but I just want to make sure: when passing a custom rate matrix to ace, does the transition from state i to state j go in position [i,j] of the matrix, or in position [j,i]? If this is just a typo in the wiki, who do I contact about correcting it? Thanks! Rosemary _______________________________________________ R-sig-phylo mailing list - [email protected] https://stat.ethz.ch/mailman/listinfo/r-sig-phylo Searchable archive at http://www.mail-archive.com/[email protected]/
