NEWS

corr2D 1.0.3 (2022-07-14)

Moved package profr and xtable to Suggests. They are optionally used to check corr2ds performance and are thus not essential to the package.

Moved package rgl to Suggests as the future of OpenGL is uncertain. rgl is used as an optional preview in corr2d and is thus not essential to the package.

Added publication in Journal of Statistical Software as preferred citation. The publication is a good start for newcomers to the field of 2D correlation spectroscopy (in R)!

Added internal 'testthat' tests to corr2D to easier check correlation results during package development.
Added a default print() method for corr2d objects (plot.default() for now).
Changed plot_corr2d() to use plot() instead of plot.default() to enable method dispatch.
Fixed a small error with the calculation of reference spectrum 2 (Ref2) if reference spectrum 1 (Ref1) is not present (Thanks to Bettina Gruen).

Added 2T2D correlation analysis to the package. Function uses the approach described by I. Noda (2018) <DOI: https://doi.org/10.1016/j.molstruc.2018.01.091>.
A freshly published paper at The Journal of Statistical Software was added as a vignette. This will be updated in the future to also include descriptions of new functions.
Added citation referencing the published paper at The Journal of Statistical Software.

Added 2D codistribution analysis to the packages functions. Function uses the approach described by I. Noda (2014) <DOI: https://doi.org/10.1016/j.molstruc.2014.01.024>.
Rewrote the function plot_corr2d() for better control about the plot appearance. Introduced the graphical parameters "lwd", "lwd.axis", "lwd.spec", "col", "col.axis", "col.lab", "cex.axis", "cex.lab", "cex.leg", "font.axis" and "font.lab" which are/are derived from "par". "at.xaxs"/"at.yaxs" and "label.xaxs"/"label.yaxs" allow control over the axes ticks and their labels. "line.xlab"/"line.ylab" control the position of the axes label.

Rewrote the function plot_corr2d(): The graphical parameters specified at "..." are now partly transferred to all parts of the plot, not just the main screen.
Cleaned up the image.plot() code inside plot_corr2d().
Fixed a bug which prevented the use of an individual color palette in plot_corr2din3d().
Changed the default color palette in plot_corr2din3d() from fields::timcolors() to colorspace::diverge_hcl(). This change should help to improve the default graphics quality.

Corrected the default value for the normalization factor. It's now the number of perturbation variables which is NROW() of Mat1.
Added error massages to all functions.
Rewrote the function sim2ddata() to allow to set C and Camp NULL. In that case only the first order reaction A -> B will be simulated and sampled.

Added the function sim2ddata to simulate artificial data.
Fixed the plotting of the reference spectra on the 2D correlation spectra. Before the reference spectra used an arbitrary x-axis. Now they use their correct spectral variables for plotting. 2D correlation peaks should now align with their 1D counterparts in the reference spectra.
Rewrote the interpolation of the perturbation variables to no longer use the ineffective cbind in a for loop. Instead apply and sapply are now used instead.
The interpolated perturbation variables are now saved when the perturbation values get interpolated.
Cleaned up some left/right confusions in the documentation of plot_corr2d.

A new plot after a call of plot_corr2d() is plotted inside the main part. It's a consequence of keeping the split.screen of the main part active for interactive data readout and manipulation.