Online multi-layer FDR control

2025-06-03

AI Summary

Controlling both within-group and global false discovery rates (FDR) simultaneously in streaming multi-layer hypothesis testing—e.g., real-time identification of effective nucleic acid–nanocarrier combinations in RNA nanocapsule studies with dual stratification by nucleic acid type and delivery vehicle—remains an open challenge. Method: We propose the first online multi-layer FDR control framework, integrating adaptive p-value weighting with hierarchical dependency modeling. Leveraging martingale theory, we provide a rigorous proof that the method guarantees modified FDR (mFDR) ≤ α at every layer under arbitrary dependency structures among hypotheses. Contribution/Results: Compared to conventional single-layer online methods, our approach significantly improves statistical power and real-time responsiveness in simulations, while accommodating both nested and parallel grouping architectures. The framework establishes a provably valid, deployable paradigm for online multiple testing in cross-dimensional biomedical decision-making.

Technology Category

• Machine Learning: Online Learning & Bandits
• Reasoning under Uncertainty: Stochastic Optimization
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

When hypotheses are tested in a stream and real-time decision-making is needed, online sequential hypothesis testing procedures are needed. Furthermore, these hypotheses are commonly partitioned into groups by their nature. For example, the RNA nanocapsules can be partitioned based on therapeutic nucleic acids (siRNAs) being used, as well as the delivery nanocapsules. When selecting effective RNA nanocapsules, simultaneous false discovery rate control at multiple partition levels is needed. In this paper, we develop hypothesis testing procedures which controls false discovery rate (FDR) simultaneously for multiple partitions of hypotheses in an online fashion. We provide rigorous proofs on their FDR or modified FDR (mFDR) control properties and use extensive simulations to demonstrate their performance.

Problem

• Online FDR control for sequential hypothesis testing
• Multi-layer FDR control for grouped hypotheses
• Real-time decision-making with simultaneous FDR control

Innovation

• Online sequential hypothesis testing procedures
• Multi-layer false discovery rate control
• Rigorous FDR and mFDR control proofs