Changes in version 0.12                         

    o   Basic multitrait interface  (#30)

    o   Default values for initial (co)variances are now a function of empirical
	(co)variances

    o   Interface for using **weights** for residual variance  (#77)

    o   Vignette: Heterogeneous variances

                        Changes in version 0.11                         

    o   Install breedR from package repository avoiding package devtools

    o   New vignette: Computing Heritability

    o   Access to inverse Average-Information matrix

    o   Automatic computation of heritability for genetic models and method 'ai'

    o   remlf90() interface to PROGSF90 OPTIONS

    o   Much more exhaustive testing ensures greater stability.

    o   New datasets Douglas (multi-site) and Larix (repeated measurements)

    o   Separate R-package from BLUPF90 binaries. Exploit 64bit architectures.

    o   Several bugfixes. Notably, the empirical variogram computation and memory issues with the AR model.

                        Changes in version 0.10                         

    o   Full internal refactoring (with few 'visible' changes in the interface)

    o   `model.matrix()` no longer separates `fixed` and `random` effects.
	It returns the same structure as `ranef()` which is more intuitive.

    o   `vcov()` works now with any of the non-diagonal random effects.

                         Changes in version 0.9                         

    o   New 'generic' model component allows passing a list of arbitrary incidence and covariance/precision matrices

    o   Optimized splines model. It runs about 5x faster.

                         Changes in version 0.8                         

    o   New interface ranef() to access BLUP of random effects

    o   New interface model.matrix() to get incidence matrices

    o   Integrated vignettes into the package

                         Changes in version 0.7                         

    o   Remote computing through ssh

    o   Convenient scaling of spatial variance components

    o   Additive genetic competition model

    o   Permanent environmental competition effect

    o   Simulation infrastructure

                         Changes in version 0.6                         

    o   Visualization and diagnostic tools

    o   Prediction in unobserved locations

                         Changes in version 0.5                         

    o   Customizable initial variance components

    o   breedR.option infrastructure

                         Changes in version 0.4                         

    o   AR1xAR1 spatial model

    o   Eucalyptus Globulus dataset and demo

                         Changes in version 0.3                         

    o   Pedigree checking, completing, sorting and recoding

                         Changes in version 0.2                         

    o   Unstructured Random effects

    o   Simple simulation of spatially structured effects

    o   Spatial module following Cappa and Cantet (2007) splines model

                         Changes in version 0.1                         

    o   Basic wrapper to Mistalz's (AI)REMLF90

    o   Convenient 'metagene' class to represent and manipulate simulated datasets