Plotindiv color issue

1

Could you kindly clarify the reasoning behind the color inversion between the data points, confidence interval circles, and prediction background in the EWS and RMS groups, as shown in Figures 1 and 2? I also noticed that this does not occur in the BL and NB groups—would you be able to explain why? Thank you for your time and help!

plotIndiv(final.plsda.srbct, ind.names = FALSE, legend=TRUE,
comp=c(1,2), ellipse = TRUE,
title = ‘PLS-DA on SRBCT comp 1-2’,
X.label = ‘PLS-DA comp 1’, Y.label = ‘PLS-DA comp 2’)
Figure1

517716f640fda224ce7abec2be8d129_compressed
517716f640fda224ce7abec2be8d129_compressed867×780 11.6 KB

plotIndiv(final.plsda.srbct, comp = 1:2, group = srbct$class,
ind.names = FALSE, title = ‘Mahalanobis.dist’,
legend = TRUE, background = background.max)
Figure2

image
image867×780 10.5 KB

2

Hi @Ande2049,

This bug has been fixed in the latest version of mixOmics which you can install using Github now devtools::install_github("mixOmicsTeam/mixOmics") or will also be available on Bioconductor from the end of next week :slight_smile:
Cheers,
Eva