| Calling Sequence
| RBFS_Tree(t,Dist,Var)
|
| Parameters
| | Name | Type | Description |
|
| t
| Tree | input distance Tree |
| Dist
| matrix(numeric) | distance matrix used to build Tree |
| Var
| matrix(numeric) | variances of the distances |
| 'Top' = posint | (default=1) number of best trees to return |
|
| Return Type
| set([numeric, Tree]) |
| Synopsis
| RBFS_Tree is a method for improving distance phylogenetic
trees using heuristics.
The first type of heuristics, called Reduce Best Fitting Subtree (RBFS)
selects a set of subtrees which are highly consistent and their fit is
of good quality, replaces them with a single leaf and attempts to
optimize the reduced tree.
The second heuristic chooses, from different trees, subtrees which are
on the same set of leaves and tries to graft them together hoping that
the resulting tree is better.
RBFS_Tree returns a set of pairs: [DimensionlessFit,Tree].
The number of trees returned can be changed with the optional parameter
Top=n.
The trees returned are the ones which have the highest quality (lowest
DimensionlessFit value). |
| See also
| BootstrapTree, ComputeDimensionlessFit, GapTree, Leaf, LeastSquaresTree, PhylogeneticTree, SignedSynteny, Synteny, Tree |