swMATH ID: 
41983

Software Authors: 
Liang, Jane W.; Sen, Śaunak

Description: 
Sparse matrix linear models for structured highthroughput data. Recent technological advancements have led to the rapid generation of highthroughput biological data which can be used to address novel scientific questions in broad areas of research. These data can be thought of as a large matrix with covariates annotating both its rows and columns. Matrix linear models provide a convenient way for modeling such data. In many situations, sparse estimation of these models is desired. We present fast, general methods for fitting sparse matrix linear models to structured highthroughput data. We induce model sparsity using an ({L_1}) penalty and consider the case when the response matrix and the covariate matrices are large. Due to data size, standard methods for estimation of these penalized regression models fail if the problem is converted to the corresponding univariate regression scenario. By leveraging matrix properties in the structure of our model, we develop several fast estimation algorithms (coordinate descent, FISTA and ADMM) and discuss their tradeoffs. We evaluate our method’s performance on simulated data, extit{E. coli} chemical genetic screening data and two extit{Arabidopsis} genetic datasets with multivariate responses. Our algorithms have been implemented in the Julia programming language and are available at url{https://github.com/senresearch/MatrixLMnet.jl}. 
Homepage: 
https://projecteuclid.org/journals/annalsofappliedstatistics/volume16/issue1/Sparsematrixlinearmodelsforstructuredhighthroughputdata/10.1214/21AOAS1444.short

Source Code: 
https://github.com/senresearch/MatrixLMnet.jl

Dependencies: 
Julia 
Keywords: 
gradient descent;
proximal gradient algorithms;
ADMM;
FISTA;
Julia;
Lasso

Related Software: 
glmnet;
sma;
GitHub;
data.table;
R/qtl;
selectiveInference;
Julia;
Mutoss;
hglasso;
limma;
MESS;
fda (R);
R

Cited in: 
1 Publication
