lrcde.permutations: lrcde.permutations

In ERGlass/lrcde.dev: Linear Regression Cell Type-Specific Differential Expression Power Analysis

Description

Call this function to run entire functionality of LRCDE.

Usage

lrcde.permutations(het.sub, cell.props, groups, medCntr = FALSE,
  stdz = FALSE, nonNeg = TRUE, method = "dual", direction = "two.sided",
  n.perms = 100, cell.p = 1, truth)

Arguments

het.sub	Should be samples by genomic site (rows by columns). The samples by genomic measures heterogeneous observations matrix.
cell.props	Should be samples by cell types (rows by columns). The relative cell proportions per sample.
groups	A vector of 1's and 2's indicating group membership per sample. Should align with samples in het.sub and cell.props.
medCntr	Boolean indicating whether to mean center differential expression estimates.
stdz	Boolean indicatin whether to scale differential expression estimates with their pooled adjusted standard deviation
nonNeg	Boolean indicating whether to force cell type-specific estimates to be non-negative (TRUE) or not (FALSE).
method	Only "dual" is implemented in this version. This should be one of "dual" (csSAM method), or "ridge". Default is "dual". Specifies which type of regression deconvolution to perform.
direction	Should be one of "two.sided", "up", or "down". Which direction to test for cell type-specific expression changes.
n.perms	Number of permutations of group labels to perform
cell.p	The simulation differentiated target cell.
truth	The truth indicator vector of which simulated "genes" in the target cell are differentiated between cases and controls.

Value

List containing data.frame (total.frame) of analysis results and a list of parameter values supplied to lrcde function (arg.used).

Author(s)

Edmund R Glass, [e-mail], Mikhail G Dozmorov, [e-mail]

References

https://github.com/ERGlass/lrcde.dev

ERGlass/lrcde.dev documentation built on May 6, 2019, 3:09 p.m.