Mendelian Randomization Methods for Causal Inference: Estimands, Identification and Inference

2025-09-14

AI Summary

Mendelian randomization (MR) faces methodological challenges including invalid or weak instrumental variables (IVs), integration of multi-source data, and adaptation to high-dimensional omics. To address these, we develop a unified analytical framework centered on causal estimands, accommodating family- and population-based study designs, individual- and summary-level data, and single- and two-sample MR settings. We propose novel strategies to detect and correct for bias arising from invalid and weak IVs, and integrate state-of-the-art software tools to enable joint modeling of multiple instruments. Our approach advances robust causal inference with genetic instruments under complex data structures and extends MR methodology to high-dimensional omics. Empirical validation using UK Biobank and Alzheimer’s disease proteomics data demonstrates improved estimation accuracy and reliability. This work provides a theoretically rigorous yet practically implementable MR analysis framework for biomedical and public health research.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Intelligent Robots: Multi-Robot Systems

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Mendelian randomization (MR) has become an essential tool for causal inference in biomedical and public health research. By using genetic variants as instrumental variables, MR helps address unmeasured confounding and reverse causation, offering a quasi-experimental framework to evaluate causal effects of modifiable exposures on health outcomes. Despite its promise, MR faces substantial methodological challenges, including invalid instruments, weak instrument bias, and design complexities across different data structures. In this tutorial review, we provide a comprehensive overview of MR methods for causal inference, emphasizing clarity of causal interpretation, study design comparisons, availability of software tools, and practical guidance for applied scientists. We organize the review around causal estimands, ensuring that analyses are anchored to well-defined causal questions. We discuss the problems of invalid and weak instruments, comparing available strategies for their detection and correction. We integrate discussions of population-based versus family-based MR designs, analyses based on individual-level versus summary-level data, and one-sample versus two-sample MR designs, highlighting their relative advantages and limitations. We also summarize recent methodological advances and software developments that extend MR to settings with many weak or invalid instruments and to modern high-dimensional omics data. Real-data applications, including UK Biobank and Alzheimer's disease proteomics studies, illustrate the use of these methods in practice. This review aims to serve as a tutorial-style reference for both methodologists and applied scientists.

Problem

• Addressing unmeasured confounding and reverse causation in causal inference
• Overcoming methodological challenges like invalid and weak instruments
• Extending Mendelian randomization to high-dimensional omics data settings

Innovation

• Genetic variants as instrumental variables
• Strategies for detecting invalid instruments
• Methods for high-dimensional omics data