Tractor-Mix Tool Identifies Disease-Linked Genes Across Diverse Ancestries

Genetic research has long faced a significant bottleneck: the reliance on relatively homogeneous study populations. By assuming participants are unrelated or share a single ancestral background, many traditional statistical methods struggle to produce accurate results when applied to more diverse, real-world groups. A new computational approach, Tractor-Mix, aims to resolve this issue by directly modeling the complex structure of human ancestry and familial ties within genetic datasets.

Dr. Elizabeth Atkinson on Modeling Genetic Complexity

The development, led by researchers at Baylor College of Medicine and the Texas Children’s Duncan Neurological Research Institute, addresses the loss of data that occurs when researchers must exclude participants to maintain statistical rigor. Senior author Dr. Elizabeth Atkinson, an investigator at the Duncan NRI and assistant professor at Baylor, noted that existing methods were often too rigid for modern, interconnected populations.

“Many of the statistical methods used in genetic research today were developed using relatively homogeneous populations or assume study participants are unrelated,” said senior author Dr. Elizabeth Atkinson, assistant professor in the Department of Molecular and Human Genetics at Baylor and an investigator at the Duncan NRI. “As genetic studies grow, they are capturing more of the real structure of human populations, including ancestry from multiple populations and family relationships among participants. Tractor-Mix gives researchers a way to model that structure directly, so more participants can be included in analyses while maintaining the statistical rigor needed to identify disease-linked genetic signals.”

Dr. Elizabeth Atkinson, assistant professor in the Department of Molecular and Human Genetics at Baylor and an investigator at the Duncan NRI

By allowing scientists to account for ancestry and family connections simultaneously, the tool helps separate genetic signals that previously appeared blurred. This increased precision is intended to improve the understanding of disease biology, particularly for conditions like diabetes, cancer, and various neurological disorders.

Testing Tractor-Mix with UK Biobank and Mexico City Data

To validate the effectiveness of Tractor-Mix, the research team applied the tool to large-scale genetic datasets, including the UK Biobank and the Mexico City Prospective Study. The latter is noted for its high degree of mixed ancestry and extensive family relationships, making it an ideal environment to test the method’s capabilities.

The results showed that Tractor-Mix not only confirmed known genetic associations but also revealed new signals that conventional GWAS (genome-wide association studies) methods had previously overlooked. One specific analysis successfully identified a previously undetected genetic region linked to body mass index (BMI). According to the researchers, this ability to pinpoint which ancestral background drives a specific signal could be transformative for future precision medicine efforts.

Collaborative Development and Scientific Impact

The study, published in Nature Genetics on July 22, 2026, was a broad collaborative effort. First author Taotao Tan worked alongside researchers from numerous international institutions, including Yale University, the Broad Institute, Harvard Medical School, Massachusetts General Hospital, the University of Oxford, and the National Autonomous University of Mexico.

Funding for the project was provided by the National Institutes of Health, the Caroline Wiess Law Fund for Research in Molecular Medicine, and the ARCO Foundation Young Teacher-Investigator Fund at Baylor College of Medicine. As human populations continue to become more interconnected, the researchers believe that adapting statistical methods is essential for the future of genetic discovery.

“Our goal is to make genetic discovery both more accurate and more complete,” Atkinson said. “By developing methods that better reflect real-world populations, we improve our ability to understand disease biology and move closer to precision medicine that benefits patients of all genetic backgrounds.”

Dr. Elizabeth Atkinson, Baylor College of Medicine and Duncan NRI

The team hopes that as the tool is adopted by the wider scientific community, it will lead to the discovery of disease-linked genes that have remained hidden due to the limitations of older, less inclusive statistical models.