New Statistical Tool Reveals Hidden Genetic Pathways in Complex Diseases

Complex diseases like Parkinson’s, breast cancer, and high cholesterol often differ widely in their causes, even when patients share the same diagnosis. Some individuals develop disease due to thousands of common genetic variants, while others are affected by a single rare mutation. Traditional large-scale genetic studies have obscured these differences by averaging genetic effects across all patients. Researchers have now introduced a statistical approach that detects hidden genetic drivers and subgroups patients by the true biological causes of their illness. This opens the door to clinical testing and tools that can match patients to therapies based on the actual mechanism driving their disease.

Developed by researchers at Rice University (Houston, TX, USA), Baylor College of Medicine (Houston, TX, USA), and collaborators, the method called the Causal Pivot is an extension of Causal Analysis, a branch of statistical science, and introduces new computational tools for studying complex diseases. The approach leverages polygenic risk scores (PRS), which summarize the combined effects of common genetic variants, as a pivot point to test for additional rare, harmful mutations.

The Causal Pivot formalizes the idea that if a rare variant drives disease in some people, those carriers will have lower PRS values compared to non-carriers. This creates a rigorous statistical test for identifying rare variant–driven subgroups and estimating their effect size. Unlike traditional genome-wide association studies, the method can work with cases-only designs, a major advantage in studies where healthy controls are scarce. Safeguards against confounding factors, such as ancestry, further ensure accuracy across diverse populations.

In a study published in the American Journal of Human Genetics, the research team validated the method using UK Biobank data. Researchers applied it to well-studied gene-disease pairs, including LDLR in high cholesterol, BRCA1 in breast cancer, and GBA1 in Parkinson’s disease. In each case, the Causal Pivot detected clear signals that aligned with known biology, while tests with unrelated genes or harmless variants produced no false positives. The approach also revealed new insights into lysosomal storage pathways in Parkinson’s, showing how multiple rare variants can combine to drive disease.

The findings demonstrate how subgroup detection can reshape genetic medicine by allowing doctors to match patients with therapies based on mechanisms rather than symptoms. By revealing genetic “routes” into a disease, the Causal Pivot could guide more targeted genetic testing, improve clinical trial design, and provide a framework for mechanism-specific treatments. Researchers also envision the approach extending beyond genetics, using environmental exposures, biomarkers, or imaging features as pivot points to uncover other disease drivers.

“Not everyone with a complex disease gets there the same way. The Causal Pivot is designed to detect those differences and sort patients into more precise, biologically meaningful subgroups. This is a foundational step toward truly personalized genetic medicine,” said Chad Shaw, statistician at Rice University and Baylor College of Medicine, and director of Rice’s Data to Knowledge Lab.