# KANN: estimation of genetic ancestry profiles by nearest neighbor regression

## TL;DR

KANN is a fast and efficient method for estimating genetic ancestry using nearest neighbor regression, suitable for large-scale genomic studies.

## Contribution

KANN introduces a novel approach for ancestry estimation using k-nearest neighbor regression with continuous ancestry profiles.

## Key findings

- KANN's ancestry estimates align well with haplotype-based methods like SOURCEFIND in Finnish biobank data.
- KANN produces results comparable to ADMIXTURE in the 1000 Genomes Project dataset.
- KANN is computationally efficient and suitable for biobank-scale analyses.

## Abstract

State-of-the-art methods for inferring individual-level genetic ancestry are based on statistical models for haplotype data. Unfortunately, these methods are computationally demanding, making them impractical for biobank-scale analyses. In this paper, we describe KANN, an efficient k-nearest neighbor regression method for individual-level ancestry estimation with respect to predefined source populations using only principal components of genetic structure. Contrary to the existing tools that can only use reference samples with discrete source population assignment, KANN enables the use of reference samples with continuous ancestry profiles across multiple source populations. We observe that KANN’s ancestry estimates agree well with the haplotype-based method SOURCEFIND when estimating ancestry profiles across up to 10 Finnish source populations on a dataset of 18 125 Finnish samples from THL Biobank. In the 1000 Genomes Project data containing globally diverse genetic backgrounds, KANN produces highly similar results to the ADMIXTURE software. Based on our results, KANN is a promising tool for ancestry estimation in large-scale genomic studies.

Graphical Abstract