Hierarchical Modeling for Rare Event Detection and Cell Subset Alignment across Flow Cytometry Samples
Figure 1
Schematic summary illustrating the HDP model framework.
A graphical model provides a declarative representation of the HDPGMM. The figure shows a compact plate representation of the graphical model, in which plates (rounded rectangles) are used to group variables in a subgraph. Each subgraph in a plate is replicated a number of times as indicated by the label within the plate. The event in the
sample is represented by
, and the
component for the
sample is a multivariate Gaussian with proportion
, mean
and covariance matrix
.s. Hyper-parameters that can be set are
,
,
,
,
and
as described in Methods. Given the declarative graphical model, standard and GPU-accelerated MCMC sampling algorithms can be used to implement the model as previously described [16].