Some views have too many features – suggestions for dimensionality reduction?

[JanakiRamya36]

I’m using MOFA2 on bulk multi-omics data (methylation and gene expression) with ~95k CpG sites and ~20k genes across 270 samples. While creating the MOFA object, I get the error: "Some views have too many features. Please consider doing stringent feature selection before proceeding."

Is there a recommended max number of features per view and What’s the best way to reduce data dimensionality?

[gtca]

One of the ways would be to only keep some highly variable features (e.g. top few thousand genes and e.g. top 10k CpG sites).