Sparse matrix linear models for structured high-throughput data

TL;DR

This paper introduces fast, scalable methods for fitting sparse matrix linear models to high-throughput biological data, enabling efficient analysis of large structured datasets with covariates on both rows and columns.

Contribution

It develops novel algorithms leveraging matrix properties for sparse estimation in large matrix linear models, addressing computational challenges in high-dimensional data.

Findings

Algorithms perform well on simulated data.

Effective analysis of biological datasets demonstrated.

Methods are implemented in Julia for accessibility.

Abstract

Recent technological advancements have led to the rapid generation of high-throughput 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 rows and columns of this matrix. 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 high-throughput 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 trade-offs. We evaluate our method's performance on simulated data, E. coli chemical genetic screening data, and two Arabidopsis genetic datasets with multivariate responses. Our algorithms have been implemented in the Julia programming language and are available at https://github.com/senresearch/MatrixLMnet.jl.