Method question ： Bayesian Network Latent Mixture Model

[kane9530]

I read in the Bayesian Network Latent Mixture Model section that the model provides predictions on an individual cell-level basis.

I don't quite understand this statement. Does this mean that an entire cell is labelled as "amplified/neutral/deletion"? Why would this be useful, as biologically, CNVs are defined on a regional level and not a whole cell level? Would appreciate some clarity on this!