How does limma-trend work? An empirical partially Bayes perspective

TL;DR

This paper analyzes the limma-trend method for high-throughput data using an empirical partially Bayes framework, introduces a nonparametric extension, and discusses FDR control under various conditions.

Contribution

It provides a new empirical Bayes perspective on limma-trend, proposes a nonparametric generalization, and enhances understanding of FDR control in high-throughput analyses.

Findings

Limma-trend computes approximate partially Bayes p-values conditioned on residual variance.

A nonparametric extension asymptotically controls FDR under dense signals.

The framework explains why some variants like MAnorm2 may fail to control FDR.

Abstract

In high-throughput biology, it is common to fit thousands of linear regressions -- one per gene, protein, or other unit -- with very few samples per unit. Limma-trend, one of the most widely used methods in this setting, improves power by shrinking variance estimates parametrically toward a fitted curve (the trend) relating variance to a unit-level summary (e.g., average intensity, peptide count), before computing p-values and applying the Benjamini-Hochberg procedure to control the false discovery rate (FDR). We study limma-trend through the lens of empirical partially Bayes inference, a paradigm in which a prior is posited and estimated for the nuisance parameters while parameters of interest remain fixed. From this perspective, limma-trend computes approximate partially Bayes p-values that condition on the residual sample variance and the unit-level summary. The same framework explains why MAnorm2, a popular variant for ChIP-seq, can sometimes fail to control FDR. We then derive a nonparametric generalization of limma-trend that estimates the residual variance prior using nonparametric maximum likelihood. Under dense signals, this procedure asymptotically controls the FDR -- even when the trend is misspecified or inconsistently estimated. To allow the full shape of the conditional variance distribution to depend on the unit-level summary, we develop a second procedure that learns it directly.