| Title: | Bayesian Hierarchical Models for Single-Cell Protein Data |
| Version: | 1.0.0 |
| Description: | Bayesian Hierarchical beta-binomial models for modeling cell population to predictors/exposures. This package utilizes 'runjags' to run Gibbs sampling with parallel chains. Options for different covariances/relationship structures between parameters of interest. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.3.2 |
| Imports: | coda, runjags, VGAM, matlib |
| Depends: | R (≥ 3.5), rjags |
| Suggests: | knitr, rmarkdown |
| VignetteBuilder: | knitr |
| LazyData: | true |
| NeedsCompilation: | no |
| Packaged: | 2025-09-30 21:02:27 UTC; cjsakitis |
| Author: | Chase Sakitis [aut, cre], Brooke Fridley [aut] |
| Maintainer: | Chase Sakitis <cjsakitis@cmh.edu> |
| Repository: | CRAN |
| Date/Publication: | 2025-10-07 18:00:12 UTC |
Bayesian Immune Cell Abundance Model (BICAM)
Description
Bayesian Immune Cell Abundance Model (BICAM)
Usage
BICAM(
dat,
M,
adapt,
burn,
it,
thin = 1,
ran_eff = 1,
chains = 4,
cores = 4,
v0_mu_logit = 0.01,
ncov = 1,
model = "Unstr",
dis = NULL,
tree = NULL,
treelevels = NULL
)
Arguments
dat |
data frame with dataset (proper setup displayed in tutorial) |
M |
number of cell types/parameters of interest |
adapt |
number of adaptation iterations (for compiling model) |
burn |
number of burn-in iterations |
it |
number of sampling iterations (after burn-in) |
thin |
number of thinning samples |
ran_eff |
indicate whether to use random subject effect (repeated measurements) |
chains |
number of chains to run |
cores |
number of cores |
v0_mu_logit |
anticipated proportion of cell types/parameters |
ncov |
number of covariates input into the model |
model |
covariance model selection |
dis |
distance matrix for Exp. Decay model |
tree |
tree-structured covariance matrix for Tree and Scaled Tree models |
treelevels |
list of matrices for multilevel, tree-structured covariance matrix for TreeLevels model |
Value
A list of inputs and results
Examples
data(dat)
BICAM(dat,2,1500,250,250)
Example dataset: dat
Description
A sample dataset used for demonstrating the function.
Usage
dat
Format
A data frame with 10 rows and 5 columns:
- suid
Subject ID's
- total
Total number of trials
- stage
Binary predictor variable (0/1)
- M1
Count data for Marker 1
- M2
Count data for Marker 2
Source
Imported from CSV and saved as RData
Examples
data(dat)
head(dat)