A1115
Title: A general statistical framework for FDR control in feature screening of single-cell genomics
Authors: Xinzhou Ge - Oregon State University (United States) [presenting]
Abstract: Large-scale feature screening is essential in high-throughput biological data analysis, particularly for identifying genes that exhibit differential expression across various conditions. The false discovery rate is the most widely used criterion to ensure the reliability of screened features. The most famous Benjamini-Hochberg procedure for FDR control requires valid high-resolution p-values, which are, however, often hardly achievable because of the reliance on reasonable distributional assumptions or large sample sizes. We propose a general statistical framework, Clipper, for large-scale feature screening with theoretical FDR control and without p-value requirement. Extensive numerical studies have verified that Clipper is a versatile and effective tool for correcting the FDR inflation crisis in multiple bioinformatics applications. Notably, it effectively resolves the ``double dipping'' issue prevalent in single-cell genomic analyses.
