[Bioperl-l] Posterior (and other values) from BEAST/treeannotator/FigTree tree nodes

Jason Stajich jason at bioperl.org
Thu Jul 8 00:27:56 EDT 2010

Hi -

Is this just that you have internal nodes labeled with the 
bootstrap/posterior value? If so you just want the node ids for the 
internal nodes then.  The semantics of what is encoded in the internal 
nodes is not forced by the nexus format so it assumes they are ids by 
You can call $tree->move_id_to_bootstrap
to migrate the ids of internal nodes to the bootstrap option.

you can also use the -internal_node_id => 'bootstrap' option when 
initializing TreeIO objects too:

=head2 new

  Title   : new
  Usage   : my $obj = Bio::TreeIO->new();
  Function: Builds a new Bio::TreeIO object
  Returns : Bio::TreeIO
  Args    : a hash.  useful keys:
    -format : Specify the format of the file.  Supported formats:

      newick             Newick tree format
      nexus              Nexus tree format
      nhx                NHX tree format
      svggraph           SVG graphical representation of tree
      tabtree            ASCII text representation of tree
      lintree            lintree output format
    -internal_node_id : what is stored in the internal node ids,
                        bootstrap values or ids, coded as
                        'bootstrap' or 'id'

Greg Baillie wrote, On 7/7/10 7:41 AM:
> Hi.
> I'm trying to extract information (eg. posterior values) from nodes of 
> trees generated using BEAST/treeannotator/FigTree.
> The trees, which are in nexus format, seem to load OK, but when I 
> iterate through the nodes looking for posterior values (/if ( 
> $node->has_tag('posterior') ) {.../), none of the nodes seem to have 
> posteriors.
> In fact, none of the nodes have any tags (/my @tags = 
> $node->get_all_tags(); print join ',', @tags/). And when I dump an 
> entire tree using Data::Dumper, there is no sign of any of the node 
> values.
> Does TreeIO handle these values? If so, should I be looking somewhere 
> other than the tags for the values?
> Thanks.
> Greg. 

More information about the Bioperl-l mailing list