Lewis model in BEAST - MK or MKv

[Mike Lee]

Hi All,

I was just wondering if the Lewis model in BEAST is the MK model, or
the MKv model (which corrects for ascertainment bias, where only
variable characters are scored - as is often the case in morphology).
MrBayes has these corrections for when only variable, or only
parsimony-informative characters are scored.

From Alex's post below, I'm guessing it's the uncorrected MK model?

Cheers, Mike Lee



From: Alexander Alekseyenko <alexander.alekseye...@gmail.com>
Date: Fri, 23 Apr 2010 07:50:12 -0700 (PDT)
Local: Fri, Apr 23 2010 11:50 pm
Subject: Re: phylogeography
Reply to author | Forward | Print | Individual message | Show original
| Report this message | Find messages by this author
<lewisMk> is a wrapper around the GeneralSubstitutionModel, which
allows for easier specification of ordered trait substitutions. Note
however that the non-zero rates are all equal there as in the original
Lewis model.

Cheers,

Alex