Continuing the discussion from TSS normalization of microbiome data for sPLS:
Hi @Tonima ,
Apologies for the delay as we are changing over maintainers.
For longitudinal studies this is a bit tricky. The current methods in mixOmics (except for multilevel decomposition) consider each sample as independent, so it really depends on what you are focusing on in your data (e.g time effect is not important?). You can have a look at our examples in multilevel decomposition on our website. The multilevel function would do a cross-validation taking into account the repeated nature of the measurements. Multilevel – mixOmics
Of note, we are running some advanced workshops early next year focusing on longitudinal designs, if that is of interest: Workshops – mixOmics .
Kim-Anh