AlternateSVD {leapp} | R Documentation |
Alternating singular value decomposition
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 |
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 yunting.sun@gmail.com, Nancy R.Zhang nzhang@stanford.edu, Art B.Owen owen@stanford.edu