Guiding adaptive shrinkage by co-data to improve regression-based prediction and feature selection

TL;DR

This paper reviews and demonstrates guided adaptive shrinkage methods that leverage co-data to enhance regression prediction and feature selection in high-dimensional genomics data, addressing challenges in low sample size studies.

Contribution

It provides a comprehensive review of guided adaptive shrinkage techniques, compares them with other methods, and shows how to integrate co-data learners with spike-and-slab priors for improved feature selection.

Findings

Guided adaptive shrinkage improves feature selection accuracy.

Comparison shows group-adaptive shrinkage outperforms sparse group-lasso.

Integration with spike-and-slab enhances genetics study analysis.

Abstract

The high dimensional nature of genomics data complicates feature selection, in particular in low sample size studies - not uncommon in clinical prediction settings. It is widely recognized that complementary data on the features, `co-data', may improve results. Examples are prior feature groups or p-values from a related study. Such co-data are ubiquitous in genomics settings due to the availability of public repositories. Yet, the uptake of learning methods that structurally use such co-data is limited. We review guided adaptive shrinkage methods: a class of regression-based learners that use co-data to adapt the shrinkage parameters, crucial for the performance of those learners. We discuss technical aspects, but also the applicability in terms of types of co-data that can be handled. This class of methods is contrasted with several others. In particular, group-adaptive shrinkage is compared with the better-known sparse group-lasso by evaluating feature selection. Finally, we demonstrate the versatility of the guided shrinkage methodology by showing how to `do-it-yourself': we integrate implementations of a co-data learner and the spike-and-slab prior for the purpose of improving feature selection in genetics studies.