glm.init: Initializes parameter estimates

In DavidKLim/FSCseq: Feature Selection and Clustering of RNA-seq Count Data

Description

Uses glm() and phi_ml_g() functions to initialize estimates at the beginning of EM algorithm.

Usage

glm.init(j, y_j, XX, k, offsets, wts, keep)

Arguments

j	integer, gene index j
y_j	vector of length n, expression of gene j
XX	matrix of dimension nk by k+p, design matrix
k	integer, number of clusters
offsets	numeric vector of length n, subject-specific offsets
wts	matrix of dimension k by n, E step weights
keep	logical matrix of dimension k by n, whether a sample is included in calculation for cluster k

Value

list containing outputs coefs_j: vector of length k corresponding to each cluster log2 baseline, phi_g: Overdispersion estimate

DavidKLim/FSCseq documentation built on Dec. 12, 2021, 3:46 a.m.