Combining whole genome sequencing and non-adaptive group testing for large-scale ethnicity screens

Background: Estimating an individual’s ethnicity from genetic data is crucial for analyzing disease association studies, making informed medical decisions, conducting forensic investigations, and tracing genealogical ancestry. Results: This work combines non-adaptive group testing using the mathematical field of compressed sensing and standard short-read sequencing to allow an up to 4-fold increase in the number of samples in large-scale ethnicity estimates. The method requires no prior knowledge regarding the tested individuals and provides almost identical results compared to testing each individual independently. Our results are based on simulated data, and on simulations based on experimental data from the 1000 Genomes Project and the Human Genome Diversity Project. Conclusions: Our computational approach aims to reduce the costs of large-scale ancestry testing by up to 4-fold in many real-life scenarios while not compromising accuracy. We hope this method will allow more efficient large-scale ethnicity screenings.