Fixing Newick Starting Tree in SNAPP

68 views
Skip to first unread message

Paul Maier

unread,
Jun 7, 2019, 5:24:08 PM6/7/19
to beast-users
Hi,

When trying to substitute the UPGMA starting tree in SNAPP with a specified Newick tree, I run into an error.

The line I substitute is this one:
<stateNode id="tree" spec="beast.util.ClusterTree" clusterType="upgma" nodetype="snap.NodeData">

In exchange for this one:
<stateNode id="tree" spec="beast.util.TreeParser" IsLabelledNewick="true" nodetype="snap.NodeData" newick="(Taxon_A(Taxon_B,Taxon_C));">

I have read this article about BEAST starting trees, as well as several related Google Groups forums, but none of the suggested fixes has so far worked:

The error is this familiar line:
validate and intialize error: Label 'God_K101' in Newick beast.tree could not be identified. Perhaps taxa or taxonset is not specified?

Where the labels (e.g. God_K101) in the Newick tree are identical to the labels in the alignment.

Little help?

Remco Bouckaert

unread,
Jun 9, 2019, 7:29:53 PM6/9/19
to beast...@googlegroups.com
Hi Paul,

Are you perhaps mixing taxa for the sequences with those of the species tree? Note that if taxa names for the species equal that of the gene, BEAUti automatically can add a number to the species id to guarantee they are uniquely named. So, if ‘God_K101’ is the id for a sequence, possibly the species taxon can be ‘God_K1011'.

Cheers,

Remco

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To post to this group, send email to beast...@googlegroups.com.
Visit this group at https://groups.google.com/group/beast-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/8fa3347b-eef6-492b-a45a-2911f58e5ea6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Paul Maier

unread,
Jun 11, 2019, 11:59:29 PM6/11/19
to beast-users
Hi Remco,

Yes, you were exactly right. I was specifying a tree containing all taxa, not the species tree. That fixed my problem.

Thanks!
Paul

On Sunday, June 9, 2019 at 6:29:53 PM UTC-5, Remco Bouckaert wrote:
Hi Paul,

Are you perhaps mixing taxa for the sequences with those of the species tree? Note that if taxa names for the species equal that of the gene, BEAUti automatically can add a number to the species id to guarantee they are uniquely named. So, if ‘God_K101’ is the id for a sequence, possibly the species taxon can be ‘God_K1011'.

Cheers,

Remco
On 8/06/2019, at 8:21 AM, Paul Maier <mai...@gmail.com> wrote:

Hi,

When trying to substitute the UPGMA starting tree in SNAPP with a specified Newick tree, I run into an error.

The line I substitute is this one:
<stateNode id="tree" spec="beast.util.ClusterTree" clusterType="upgma" nodetype="snap.NodeData">

In exchange for this one:
<stateNode id="tree" spec="beast.util.TreeParser" IsLabelledNewick="true" nodetype="snap.NodeData" newick="(Taxon_A(Taxon_B,Taxon_C));">

I have read this article about BEAST starting trees, as well as several related Google Groups forums, but none of the suggested fixes has so far worked:

The error is this familiar line:
validate and intialize error: Label 'God_K101' in Newick beast.tree could not be identified. Perhaps taxa or taxonset is not specified?

Where the labels (e.g. God_K101) in the Newick tree are identical to the labels in the alignment.

Little help?

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages