Tractor

Summary

Tractor is a local-ancestry-informed GWAS method for admixed populations. It uses local-ancestry deconvolution to partition each individual's genotype dosage by ancestral background, then runs regression on the resulting ancestry-specific dosages, producing ancestry-specific effect sizes and P values and boosting power to detect ancestry-enriched loci. Tractor assumes unrelated samples; Tractor-Mix extends it to cohorts with relatedness via a mixed-model framework.

Method

Tractor allocates risk alleles to each local-ancestry background inferred for an individual, then performs regression (linear or logistic) on the ancestry-specific genotype dosages, optionally conditioning on total local ancestry ("hapcount") at a site and other covariates. Because it belongs to the generalized linear model family, standard use requires independent samples. The resulting ancestry-level summary statistics support downstream ancestry-aware polygenic risk scoring, meta-analysis, and fine-mapping, and can narrow credible sets within association hits.

Citations

[1] Atkinson, E. G. et al. (2021). Tractor uses local ancestry to enable the inclusion of admixed individuals in GWAS and to boost power. Nature Genetics, 53, 195–204. DOI: 10.1038/s41588-020-00766-y. Note: not independently inspected; description above is secondary, drawn from [2].

[2] Tan, T. et al. (2025). Extending Genome-Wide Association Studies to admixed cohorts with high degrees of relatedness. medRxiv preprint. DOI: 10.1101/2025.05.27.25328444. Source: 2025.05.27.25328444v1.full.md. Supports: method summary above (Introduction section, describing Tractor as prior work).