Figure 1From: BioMiCo: a supervised Bayesian model for inference of microbial community structure Plate diagram of the mixed-membership model for BioMiCo. π is the probability distribution on the possible source environments, X represents the environments, Z represents the assemblages, W represents the OTUs, θ is the prior distribution of assemblages in environments, and ϕ is the prior distribution of OTUs in assemblages. The variable α represents the concentration parameters for prior distributions, with α θ being the concentration parameter for the prior on the distribution of assemblages in environments, α ϕ the concentration parameter for the prior on the distribution of OTUs in assemblages, and α π the concentration parameter for the prior on the distribution of environments. N is the number of samples in a dataset, and N n is the number of OTUs in sample n. K is the number of environments, and L is the number of assemblages.Back to article page