blastn()
provides an interface to the command-line program blastn for nucleotide to nucleotide comparison.
combMyTree()
grafts polytomies on a phylogeny.
deleteEmptyCells()
has now a sister function identifyEmtpyCells()
that won’t delete but only identify the empty rows and columns in a sequence alignment. This is useful if there is a second matrix (e.g. with confidence scores) tied to the sequence alignment.
forceEqualTipHeights()
corrects small rounding errors in edge lengths such that the resulting phylogeny will pass ape::is.ultrametric()
.weights
in raxml()
allows to assign individual weights to each column of the alignment. It corresponds to the -a flag in RAxML (see The RAxML v8.2.x Manual for details)exec
in mrbayes()
allows to specify the name and path of the MrBayes executable explicitly. If the executable is in the search path, exec
can be missing.write.nex()
and matrixBlock
were extended to handle standard (morphological, etc.) data in a data frame. This feature including the coding of ambiguous characters was tested successfully with MrBayes.write.partioned.nex()
was removed from the package; its functionality has been integrated into write.nex()
.write.nex()
can now handle multiple DNA sequence alignments.
Argument interleave
of functions write.fas()
, write.phy()
, and write.nex()
has been renamed to block.width
for clarity.
write.partioned.nex()
will be removed soon from the package; its functionality has been integrated into write.nex()
.run = TRUE
in mrbayes()
and mrbayes.mixed()
was broken on Windows platforms. (Thanks to Liam Revell and Klaus Schliep for report and fix).mafft()
received the additional argument options
, which can be used to request options such as e.g. --adjustdirection
that are not build into the function’s interface.
This version includes a new internal function phylo2mafft()
, which does exactly the same thing as the RUBY script newick2mafft.rb
on the MAFFT website (): it converts a user-defined guide tree into a format readible by MAFFT.
delete.empty.cells()
and fillEndsWithN()
are now using Emanuel Paradis’ bit-level coding for DNA sequences, which makes them much faster.
c.genes()
has been superseeded by the cbind method for objects of class "DNAbin"
provided in the package ape; c.genes()
will be removed in one of the following versions.