statgenHTP 1.0.6
- The package has been extended by functions for modelling evolution
of the genetic signal. This extension consists of functions
fitSplineHDM()
for fitting a hierarchical data model,
predict.psHDM()
for making predictions based on this model,
and plot.psHDM()
for plotting the results. The methods used
are described in a new vignette.
- A small bug in
detectSingleOutMaize()
is fixed.
Observations with a missing value for one of the involved traits are no
longer tagged as outliers.
statgenHTP 1.0.5
statgenHTP 1.0.4
- The
removeSerieOut()
function now has an extra argument
reason allowing for restricting removal of outliers to one or more
reason the outliers where tagged.
- The
plot
function for serieOut
objects now
has an extra argument reason allowing for restricting the plotting of
outliers to one or more reason the outliers where tagged.
- The
detectSerieOut()
function is now able to handle
plotIds with irregular naming, i.e. plotIds starting with a number.
- The results of
estimateSplineParameters()
can now be
plotted in a box plot and histogram.
- In the
estimateSplineParameters()
function multiple
parameters can now be estimated at once.
- A bug in
detectSerieOut()
that caused slope outliers to
be never detected is fixed.
- A bug causing predictions to be made in
fitSpline()
for
missing values at the beginning or end of a time course is fixed.
- The
detectSerieOut()
function now checks for number of
plotIds per plant in the correct location leading to a more user
friendly error message.
statgenHTP 1.0.3
statgenHTP 1.0.2
- Plotting second derivatives of fitted splines, plotted first
derivatives. This is fixed now.
- The
detectSerieOut()
function now uses an extra
criterion for checking if time courses our outlying. See the function
documentation and vignettes for a full explanation of this new
criterion.
- The parameter
trait
in the
removeSerieOut()
function is renamed to traits
and now accepts a vector of traits that for which outlier values can be
replaced by NA
.
- Plotting the output of
detectSerieOut()
now has an
extra parameter geno.decomp
to restrict the output to
selected levels of the geno.decomp variable in the data.
- Computing the area under the curve in
estimateSplineParameters()
now allows the specification of
the time unit used on the x-axis.
- Fitting models using SpATS is now done using the option
centered = TRUE
by default.
statgenHTP 1.0.1
statgenHTP 1.0.0