search

oljagrk / jmodeltest2

Automatically exported from code.google.com/p/jmodeltest2
GNU General Public License v3.0
0 stars 0 forks source link

Extremely high rmat values #21

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
Whenever I search for a model, and the choosen model is GTR, the values for the 
rate matrix of instantaneous substitution rates [R a,b,c...f] are too high, 
reaching hundrends of thousands (excepto for one or two that will obviousy be 
one).

This values seem to be too high, according to MrBayes, which will only take 
values up to 1000.

This happened both on Mac and Windows with jModeltest 2.1.4

Original issue reported on code.google.com by dieko....@gmail.com on 16 Aug 2013 at 12:20

GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
Hi,

I have the same problem as you..too high values at both tratio rates and rmat 
rates and MrBayes doesn't accept them. Have you been able to solve the problem? 
and eventually, which was the cause for this?
I work with Windows7.

Daniela

Original comment by lucente....@gmail.com on 7 Oct 2013 at 12:19

GoogleCodeExporter commented 9 years ago 
This problem doe snot look like its solved. Any suggestions @project members?

Original comment by vjja...@gmail.com on 12 Dec 2013 at 1:31

GoogleCodeExporter commented 9 years ago 
Extremely high/low rates usually appear due to not very informative data sets. 
jModelTest takes the output from PhyML which is estimating all model 
parameters. However, you can let MrBayes estimate again this parameters giving 
the constraints of the best-fit model (i.e, equal/unequal frequencies, 
substitution scheme, ti/tv ratio, ...)

Original comment by DiegoDL84 on 12 Dec 2013 at 3:02