Variable selection via fused sparse-group lasso penalized multi-state models incorporating molecular data

TL;DR

This paper introduces a fused sparse-group lasso penalized multi-state model that effectively performs variable selection and models covariate effects across transitions using high-dimensional molecular data, with demonstrated success in simulations and AML data.

Contribution

It develops a novel FSGL penalized Cox regression framework for multi-state models, integrating transition-specific and group effects with an efficient ADMM optimization algorithm.

Findings

The method successfully selects sparse, relevant covariates in simulations.

It effectively models transition-specific effects in AML data.

Combined penalty improves model selection over global lasso.

Abstract

In multi-state models based on high-dimensional data, effective modeling strategies are required to determine an optimal, ideally parsimonious model. In particular, linking covariate effects across transitions is needed to conduct joint variable selection. A useful technique to reduce model complexity is to address homogeneous covariate effects for distinct transitions. We integrate this approach to data-driven variable selection by extended regularization methods within multi-state model building. We propose the fused sparse-group lasso (FSGL) penalized Cox-type regression in the framework of multi-state models combining the penalization concepts of pairwise differences of covariate effects along with transition grouping. For optimization, we adapt the alternating direction method of multipliers (ADMM) algorithm to transition-specific hazards regression in the multi-state setting. In a simulation study and application to acute myeloid leukemia (AML) data, we evaluate the algorithm's ability to select a sparse model incorporating relevant transition-specific effects and similar cross-transition effects. We investigate settings in which the combined penalty is beneficial compared to global lasso regularization.