This analysis confirms that the normal mixture components described previously in Figure 2 are enriched with genes from particular functional categories.
Results show that both the continuous and discrete eHMM can identify meaningful and coherent HMM mixture components that describe different properties of the data.
This result may be due to the potential inadequacy of the fixed number of mixture components to describe some or all gene sub-networks.
The Cyc sub-network was better described by mixture component one (Mix1) for most A. m. ligustica bees raised on A. m. ligustica hives (LL), meanwhile mixture component two (Mix2) better described the sub-network for A. m. mellifera bees raised on A. m. mellifera hives (MM).
This three-component mixture prior describes the three possible categories that a miRNA could be classified under: non-differentially expressed, up-regulated, or down-regulated, respectively.
This describes an infinite mixture model, where the weight of mixture component c is given by ψ c, and, for each component, the probability models for the outcome y i and the profile X i are independent, conditional on some component specific parameters Θ c and some global parameters Θ 0. In the left hand side we summarise the complete set of parameters as Θ =(Θ 0, ψ1, Θ 1, ψ2, Θ 2,…).
Pursuing this idea and extending it to multiple covariates, both categorical and continuous, would enable us to describe the joint distribution of covariates within each mixture component.
The mixture component, shown in Figure 12, is a Hierarchical HMM (HHMM) with stack transformation to a plain HMM, as described in [ 77].
The relationship between the child and parent genes in each sub-network can be described by the weighted sum of the correlations between the parent geneand child gene estimated for each mixture component.
is the assigned folded Skellam mixture component.
