error: idref treeLikelihood has not been previously declared

[Erin]

Did anyone figure out the solution to an earlier issue raised by
several folks?

I am trying to to run an analysis with 7 loci. I input a single
bifurcating starting tree (called "its") in nexus format for all 7
partitions. In BEAST, I'm getting the following error message:

"Parsing error - poorly formed BEAST file, 10million.xml: Object with
idref=its.treeLikelihood has not been previously
declared."

Like Alan and others, I had no problems saving the BEAUTi file. I've
searched the user group but cannot figure out how to fix the error.
Any help would very much be welcomed.

Thanks,
Erin
etr...@rsabg.org

[Andrew Rambaut]

Dear Erin,

It sounds like there is a bug in the XML generation of BEAUti. If you email me the XML off the group, I will take a look and file a bug report.

Andrew

> --
> 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.
>

[Pip Griffin]

Hi Erin,

yes, the problem (discovered by Christopher Blair) is that the names
of the files are duplicated for each 'treeLikelihood' command. If you
search for 'treeLikelihood' in your xml file and look carefully, you
will see this and you can just delete one copy of each name.

Pip

[Andrew Rambaut]

This issue has been logged and will be fixed for the next update:

New issue 609 by dong.w.xie: BEAUti duplicate prefix issue for multi-partition
http://code.google.com/p/beast-mcmc/issues/detail?id=609

load multi-partition and unlink all. The treelikehood section xml is broken because id has duplicated prefix but their idref does not have duplication.

<treeLikelihood id="26.26.treeLikelihood" useAmbiguities="false">

<treeLikelihood idref="26.treeLikelihood"/>

As you can see in the above, the gene name '26' is repeated in the treeLikelihood's id. It needs to be edited as Pip describes.

As this is going to affect many users, we will try to release the update as soon as we can. Now is the time to log any other bugs that are annoying you.

Andrew

[Erin Tripp]

Andrew, Pip, Thanks to all! Interestingly, I saw that suggestion several days ago, and tried to implement it. Beast still choked. I tried it again just now, and it is running. Strange, but... I'll go with it!

Thanks very much again, 

Erin

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

[cftian]

Hi Pip,

You just saved me! Thank you!

changfu

> >> For more options, visit this group athttp://groups.google.com/group/beast-users?hl=en.