False Discovery Rates in Biological Networks

TL;DR

This paper introduces a new statistical method using knockoffs for biological network analysis that effectively controls false discoveries, improving reliability in fields like neuroscience, genomics, and ecology.

Contribution

The paper presents a novel graph estimator based on knockoffs that guarantees false discovery rate control in biological network inference.

Findings

Guarantees accurate false discovery rate control

Performs well in simulations and biological data

Provides accessible R code for implementation

Abstract

The increasing availability of data has generated unprecedented prospects for network analyses in many biological fields, such as neuroscience (e.g., brain networks), genomics (e.g., gene-gene interaction networks), and ecology (e.g., species interaction networks). A powerful statistical framework for estimating such networks is Gaussian graphical models, but standard estimators for the corresponding graphs are prone to large numbers of false discoveries. In this paper, we introduce a novel graph estimator based on knockoffs that imitate the partial correlation structures of unconnected nodes. We show that this new estimator guarantees accurate control of the false discovery rate in theory, simulations, and biological applications, and we provide easy-to-use R code.