[R-sig-ME] MCMCglmm and phylogeny - huge variance problem

[Szymek Drobniak]

Hi,

it worked - estimates are much more realistic and no nasty warnings...
Thanks for your time!

Cheers,
sz.

On 7 July 2011 00:51, Jarrod Hadfield <j.hadfield at ed.ac.uk> wrote:
> ccing the list
>
> The phylogeny has internal nodes labeled, and 3 nodes are called "". If
> internal nodes are labeled, MCMCglmm is expecting them to be uniquely
> labeled.  I wonder whether this is an oversight in ape - I can't see why
> anyone would want this? Anyway...
>
> tree$node.label<-NULL
>
> should fix the problem.
>
> I would drop the redundant taxa - they just add to the computing time unless
> nodes="TIPS". nodes="ALL" is preferable.
>
> Jarrod
>
>
>
> Quoting Szymek Drobniak <geralttee at gmail.com> on Wed, 6 Jul 2011 21:45:05
> +0200:
>
>> Hi again,
>>
>> attached you'll find the data file (ref indexes studies, sp is for
>> species, genus for genus - previously we've used genera instead of
>> lower taxa, Zr is the response, mev is measurement error). In the sp
>> column genera are merged with species names - I've noticed that ape
>> removes spaces when writing tree to the file. the tree file is the one
>> with branch lengths generated by compute.brlen() and it's in Newick
>> format. It's a bit redundant (there are taxa not included in the
>> dataset but I assumed it's not a problem as it wouldn't in cas of an
>> animal model).
>>
>> Thanks for your help!
>> Cheers,
>> sz.
>>
>> On 6 July 2011 17:30, Jarrod Hadfield <j.hadfield at ed.ac.uk> wrote:
>>>
>>> Hi,
>>>
>>> There definitely seems to be a problem but its hard to figure out what
>>> it is. Could you attach the tree (and data)?
>>>
>>>
>>> Jarrod
>>>
>>>
>>> On Wed, 2011-07-06 at 17:27 +0200, Szymek Drobniak wrote:
>>>>
>>>> Hi,
>>>>
>>>> I tried compute.brlen but this works fine and computes branch lengths
>>>> without any problem. I've used such trees before (without branch
>>>> lengths, just topology) and they worked. Is it possible that this
>>>> "duplicated levels" thing messes with everything?
>>>>
>>>> sz.
>>>>
>>>> On 6 July 2011 17:11, Jarrod Hadfield <j.hadfield at ed.ac.uk> wrote:
>>>> > Hi,
>>>> >
>>>> > There seems to be something amiss with the tree. Do you expect there
>>>> > to
>>>> > be branch lengths? Do you get warning messages outside of MCMCglmm if
>>>> > you use compute.brlen on the phylogeny?
>>>> >
>>>> > Jarrod
>>>> >
>>>> >
>>
>>
>> --
>> Szymon Drobniak || Population Ecology Group
>> Institute of Environmental Sciences, Jagiellonian University
>> ul. Gronostajowa 7, 30-387 Kraków, POLAND
>> tel.: +48 12 664 52 19 fax: +48 12 664 69 12
>>
>> www.eko.uj.edu.pl/drobniak
>>
>
>
>
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>
>



-- 
Szymon Drobniak || Population Ecology Group
Institute of Environmental Sciences, Jagiellonian University
ul. Gronostajowa 7, 30-387 Kraków, POLAND
tel.: +48 12 664 52 19 fax: +48 12 664 69 12

www.eko.uj.edu.pl/drobniak