Filtering features for classification using Diablo

Analysis
1

Hello, I am new to MixOmics and try to use Diablo to analyse multiomics data from RNA, microRNA and proteomics, attempting to identify a classifier of disease. I have not found a good examples of how to filter the dataset for features which have best classifier performance. On Chapter 2 Let’s get started | mixOmics vignette I found a hint to sections of PLSDA and PLS but there X is required to be numeric matrix, while in my case it would be a list. Could you please send me or point me to an example as to how to narrow down the features in my dataset for best classification performance, and how to save them?
Thank you so much for your consideration.

2

Hi @Andreas, there is a tutorial on this page: Case study: TCGA | mixOmics. Let me know if you have any questions.

  • Christopher
3

Thank you so much and thank you for this excellent tool.
Andreas