Table 1.

Fixed parameters in the simulation design

Parameter	Description
S	Number of CpGs chosen to be associated with phenotype in simulation
μ k	Mean of the cell-type-specific DMS effects for cell type k, over all S DMSs
σ k	Standard deviation of cell-type-specific DMS effects for cell type k, over all S DMSs
σ jk	Variability of individual deviations at probe j in cell type k
${\alpha }^{\left(0\right)}={\left({\alpha }_1^{\left(0\right)},{\alpha }_2^{\left(0\right)}\right)}^{\top }$	Expected proportion of the mixture from cell types 1 and 2 when the phenotype z i is zero.
${\alpha }^{\left(Z\right)}={\left({\alpha }_1^{\left(Z\right)},{\alpha }_2^{\left(Z\right)}\right)}^{\top }$	Average cell-type mixture proportions for cell types 1 and 2 for subjects with phenotype level Z (continuous or binary)
ρ	Precision of simulated cell mixture distributions. A greater value corresponds to more clearly defined differences in cell-type proportions with respect to the phenotype