Mapping beyond diseases: Controlled variable selection for secondary phenotypes using tilted knockoffs

2025-08-25

AI Summary

Nonuniform sampling—common in biomedical case-control studies—induces collider bias, leading to spurious associations in secondary phenotype analyses; existing high-dimensional variable selection methods fail to control the false discovery rate (FDR) under such biased sampling. To address this, we propose a robust variable screening framework tailored for biased samples: leveraging selection probabilities, we construct a tilted population distribution and generate knockoff variables satisfying the model-X knockoff condition, thereby rectifying FDR inflation in nonuniformly sampled data. Our method integrates conditional independence testing with tilted distribution modeling, ensuring rigorous FDR control and high statistical power in high-dimensional settings. Extensive simulations and real case-control data analyses validate its effectiveness, successfully identifying genetic variants associated with endophenotypes.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Graphical Models
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Researchers in biomedical studies often work with samples that are not selected uniformly at random from the population of interest, a major example being a case-control study. While these designs are motivated by specific scientific questions, it is often of interest to use the data collected to pursue secondary lines of investigations. In these cases, ignoring the fact that observations are not sampled uniformly at random can lead to spurious results. For example, in a case-control study, one might identify a spurious association between an exposure and a secondary phenotype when both affect the case-control status. This phenomenon is known as collider bias in the causal inference literature. While tests of independence under biased sampling are available, these methods typically do not apply when the number of variables is large. Here, we are interested in using the biased sample to select important exposures among a multitude of possible variables with replicability guarantees. While the model-X knockoff framework has been developed to test conditional independence hypotheses with False Discovery Rate (FDR) control, we show that its naive application fails to control FDR in the presence of biased sampling. We show how tilting the population distribution with the selection probability and constructing knockoff variables according to this tilted distribution instead leads to selection with FDR control. We study the FDR and power of the tilted knockoff method using simulated examples, and apply it to identify genetic underpinning of endophenotypes in a case-control study.

Problem

• Selecting important exposures with replicability guarantees in biased samples
• Controlling false discovery rate when sampling is not uniformly random
• Identifying genetic underpinnings of endophenotypes in case-control studies

Innovation

• Tilted knockoffs for biased sampling correction
• FDR control via tilted distribution construction
• Genetic association selection with replicability guarantees