Groups and coalescent nodes

[Matt Pennell]

Hi,

I am a fairly new BEAST user and I have been using it to generate BSPs for different populations using mtDNA. The problem I am having is that for two of my alignments, BEAST cannot run the xml file as it says there are "more groups than coalescent nodes". I have gone through the xml file and compared it to a xml file from another alignment that ran properly and did not notice any difference. I am using a Coalescent: Bayesian Skyline Tree Prior. Does anyone have idea what could be going on here? Is it possibly a problem with my alignment? Any suggestions on how I can fix this issue would be greatly appreciated.

thanks,
matt pennell

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To post to this group, send email to beast...@googlegroups.com.
To unsubscribe from this group, send email to beast-users...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/beast-users?hl=en.

[alexei]

The group size is the number of steps in the population size function.
In the BSP you can't estimate more population size changes than there
are coalescent intervals in your tree. So the alignments that are not
working will be the ones with a small number of sequences. This can be
fixed by reducing the number of groups from the default of 10 to less
than or equal to n-1, where n is the number of sequences in the
alignment.

FYI: Estimating highly parametric demographic functions from single
gene alignments of just a handful of sequences is not likely to be
very illuminating. For the single-population coalescent, sequencing
more individuals will help, but there is very little return for
sequencing more than 20-40 individuals. Longer sequences are better as
long as they are completely linked (i.e. the mitochondrial genome). By
far the best thing to do is sequence multiple independent loci and use
a method that can combine the information across multiple loci (like
EBSP).

[matt pennell]

Thank you very much for your help.

I know that using multiple loci and an EBSP is definitely the way to
go. However, I am trying to look at population dynamics of
parthenogenetic (asexual) diploid organisms. As they are do not
undergo meiosis, there is no recombination between any loci and the
entire genome is effectively linked. I have some interesting questions
to ask about these populations but am not sure how I could multiple
loci to look at demographic histories other than just to place the
sequences end to end. I do not know any way around this problem but I
would be really interested to hear any ideas people had about this.

thanks again,
matt