mind: The Multi-measure INdividual Deconvolution (MIND) algorithm

In randel/MIND: Using Tissue Expression to Estimate Sample-/Subject- And Cell-Type-Specific Gene Expression via Deconvolution

The Multi-measure INdividual Deconvolution (MIND) algorithm

Description

It calculates the empirical Bayes estimates of subject- and cell-type-specific gene expression, via a computationally efficient EM algorithm.

Usage

mind(X, W, maxIter = 100, tol = 0.001, verbose = F, ncore = 4)

Arguments

X	bulk gene expression (gene x subject x measure).
W	subject-specific cell type fraction (subject x measure x cell type).
maxIter	maximum number of iterations for the EM algorithm.
tol	tolerance level of absolute relative change of the log-likelihood to stop the EM algorithm.
verbose	logical, to print the detailed information for each iteration: iter (the iteration number), logLike_change, sigma2_e, mean(diag(Sigma_c))).
ncore	number of cores to run in parallel

Value

A list containing the output of the EM deconvolution algorithm

A	the deconvolved cell-type-specific gene expression (gene x cell type x subject).
mu	the estimated profile matrix (gene x cell type).
iter	the number of iterations used in the EM algorithm.
Sigma_c	the covariance matrix for the deconvolved cell-type-specific expression (cell type x cell type).
sigma2_e	the error variance.
loglike	the log-likelihood for each EM iteration.
var_A	the posterior covariance matrix for A (vectorized covariance matrix by subject).

References

Wang, Jiebiao, Bernie Devlin, and Kathryn Roeder. "Using multiple measurements of tissue to estimate subject-and cell-type-specific gene expression." Bioinformatics 36.3 (2020): 782-788. https://doi.org/10.1093/bioinformatics/btz619

Examples

data(example)

deconv = mind(X = example$X, W = example$W, ncore = 2)

randel/MIND documentation built on May 6, 2023, 7:45 a.m.