Overlay PCA or OPLS loadings with spectra

Plotting overlayed NMR spectra. This function is based on ggplot2, a high-level plotting R package. For high ppm ranges computation time is relatively, so the range of input argument shift should be as small as possible. List argument an must have the first element define, even if it is only a single value. If colour and line width is specified, then at least one list elements of an must have the same length as X.

specload(
  mod,
  shift = c(0, 10),
  an,
  alp = 0.3,
  size = 0.5,
  pc = 1,
  type = "Backscaled",
  title = "",
  r_scale = FALSE
)

Arguments

mod

PCA or OPLS model generated via MetaboMate package functions.
shift

ppm region to visualise.
an

List with one to three elements specifying facetting, colour and linetype (see Details).
alp

Alpha value for spectral lines.
size

plot line width.
pc

index of principal component to visualise, set to 1 if input model is OPLS
type

Type of loadings visualisation, either 'Statistical reconstruction' or 'Backscaled' (see Details).
title

Plot title.
r_scale

logical, adjust limits of color gradient to 0 and 1 (only applies for argument type='stat reconstruction')

Value

ggplot2 plot object

Details

OPLS: If type='Statistical reconstruction' the function calculates the covariance (y axis) and Pearson's correlation (colouring) of the predictive OPLS scores with each X variable (x axis is ppm variable). If type='Backscaled' the OPLS loadings are backscaled with X feature standard deviations. Results are plotted over ppm, coloured according to OPLS model weights. Often, the latter method visualises model importance more robust due to the presence of false positive correlations. PCA: Function always calculates the statistical reconstruction.

NA

See also

Other NMR: alignSegment(), bline(), excise1d(), get.idx(), lw(), matspec(), noise.est(), normErectic(), plotload(), ppick(), read1d_raw(), read1d(), spec(), stocsy1d_metabom8-class, stocsy(), storm()

Author

Torben Kimhofer tkimhofer@gmail.com

torben.kimhofer@murdoch.edu.au

Examples

data(covid)
model=pca(X)

#> Too many number of components, setting pc to 2
specload(model, shift=c(1,2), an=list(an$type), pc=1, alp=0.8)