AlternateSVD: Alternating singular value decomposition

In leapp: latent effect adjustment after primary projection

Description

The algorithm alternates between 1) computing latent loadings u and latent variable v and 2) estimating noise standard deviation for each of the N genes.

Usage

AlternateSVD(x, r, pred = NULL, max.iter = 10, TOL = 1e-04)

Arguments

x	an N by n data matrix
r	a numeric, number of latent factors to estimate
pred	an n by s matrix, each column is a vector of known covariates for n samples, s covariates are considered, default to NULL
max.iter	a numeric, maximum number of iteration allowed, default to 10
TOL	a numeric, tolerance level for the algorithm to converge, default to 1e-04

Value

sigma	a vector of length N, noise standard deviations for N genes
coef	an N by s matrix, estimated coefficients for known covariates
uest	an N by r matrix, estimated latent loadings
vest	an n by r matrix, estiamted latent factors

Author(s)

Yunting Sun [e-mail], Nancy R.Zhang [e-mail], Art B.Owen [e-mail]

leapp documentation built on May 2, 2019, 2:12 p.m.