Supplementary website for

So you think you can PLS-DA?
Ruiz-Perez, Guan, Madhivanan, Mathee, Narasimhan

Daniel Ruiz-Perez Bioinformatics Research Group (BioRG) School of Computing and Information Sciences Florida International University Miami, Florida	Haibin Guan Bioinformatics Research Group (BioRG) School of Computing and Information Sciences Florida International University Miami, Florida	Purnima Madhivanan Department of Epidemiology School of Public Health Florida International University Miami, Florida	Kalai Mathee Herbert Wertheim College of Medicine Florida International University Miami, Florida	Giri Narasimhan Bioinformatics Research Group (BioRG) School of Computing and Information Sciences Florida International University Miami, Florida

This website contains all the figures from the main paper and the supplementary figures. The caption of every figure provides with a link that allows them to be rotated and interacted with in a 3D fashion.

Performance for the linearly separable points model

Figure 3. Performance for linearly separable points model, varying samples and noise (VIEW 3D MODEL).	Figure S1. Performance for linearly separable points model, varying signal and noise (VIEW 3D MODEL).	Figure S2. Performance for linearly separable points model with the cosine model, varying signal and noise (VIEW 3D MODEL).

Performance for the interval model

	Signal constrained	Noise constrained
p=3
	a) 3 Intervals constraining signal (VIEW 3D MODEL).	b) 3 Intervals constraining noise (VIEW 3D MODEL).
p=5
	c) 5 Intervals constraining signal (VIEW 3D MODEL)	d) 5 Intervals constraining noise (VIEW 3D MODEL).
p=10
	e) 10 Intervals constraining signal (VIEW 3D MODEL).	f) 10 Intervals constraining noise (VIEW 3D MODEL).
	Figure S3. Table of configurations for the interval model

Performance for the cluster model

Figure 4. Performance for the cluster model, High number of samples (VIEW 3D MODEL).

Classification cross validation accuracy for the different models

a) Accuracy for the linearly separable points model (VIEW 3D MODEL).	b) Accuracy for the Cluster model (VIEW 3D MODEL).
c) Accuracy for the Interval model constraining signal (VIEW MODEL).	d) Accuracy for the Interval model constraining noise (VIEW MODEL).
Figure S4. Experiments as a classifier

EXPERIMENTS ADDING ICA, RLDA AND SPCA

Figure S5. Performance for linearly separable points model, varying signal and noise (VIEW 3D MODEL).	Figure S6. Performance for linearly separable points model, varying samples and noise (VIEW 3D MODEL).

Figure 5. Performance signal constrained interval with p=3 (VIEW 3D MODEL).	Figure S7. Performance noise constrained interval with p=3 (VIEW 3D MODEL).

Figure S8. Performance for the cluster model, High number of samples (VIEW 3D MODEL).