Some new functionality and bug fixes

• Added function PCMTreeInsertTipsOrSingletons.
• Added possibility to specify metaI as a character string (function name) rather.
  So far, it was only possible to specify this as a function object or as the list
  resulting from calling the function on the tree, data and PCM model.
• Added an operator + for PCMTree objects. This is analogical to the operator +
  for phylo objects but maintains the edge.part and part.regime members
  (experimental).
• Removed argument parametrizations from PCMGenerateModelTypes. Now it's possible
  to specify the type of parametrization (all/default) separately for each model
  type.
• Added automatic generation of White noise model types.
• Improved man pages.

Improved APIs and docs

This package release is compatible with a major revision of the manuscript:

Mitov, V., Bartoszek, K., Asimomitis, G., & Stadler, T. Fast Likelihood Calculation for Multivariate Gaussian Phylogenetic Models with Shifts. In review.

Used for MGPM inference in mammal data and simulations

This PCMBase version contains major enhancements of the PCMTree module. Examples for the most important functions in the module are now included in the help-pages for the PCMTree* functions. This PCMBase version was used for the analysis of mammal brain and body mass measurements and simulation study included in the article "Automatic Generation of Evolutionary Hypotheses Using Mixed Gaussian Phylogenetic Models" (currently in review).

Before disentangling tree-parts from model regimes

This is the latest version of the package using the old concept of model regimes, where a regime was mapped to monophyletic or polyphyletic group of nodes on the tree.

In the versions to follow, the concept of a regime will be split into two related concepts:
In the context of a tree, we will use the word 'part' to denote a monophyletic or a paraphyletic group of nodes.
In the context of a model, we will use the word 'regime' to denote an instance of the model parameters. A suitable synonym for 'regime' would thus be 'color'. It is possible for several parts on a tree to have the same color, i.e. the same model regime.

First release on CRAN

v1.2.7

First release on CRAN