Old vs. new local ancestry inference in HCHS/SOL: a comparative study.

Area of Science:

• Population Genetics
• Genomic Ancestry Analysis
• Bioinformatics

Background:

• Hispanic/Latino populations exhibit genetic admixture from diverse ancestral groups.
• Accurate local ancestry inference is crucial for genetic association studies in admixed populations.
• Admixture mapping relies on precise estimation of ancestral origins at specific genomic loci.

Purpose of the Study:

• To evaluate the impact of updated local ancestry inference (LAI) methods on genetic analyses in Hispanic/Latino individuals.
• To compare traditional RFMix (LAI) with newer FLARE (LAI) using an updated reference panel.
• To assess the consistency of global and local ancestry estimates and their effect on admixture mapping results.

Main Methods:

• Utilized the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) dataset.
• Performed local ancestry inference using RFMix (old) and FLARE (new) with updated reference panels.
• Compared global and local ancestry correlations between inference methods.
• Analyzed admixture mapping associations derived from both inference approaches.

Main Results:

• New and old LAI methods yielded highly similar global and local ancestry estimates.
• FLARE-based admixture mapping results closely aligned with those from RFMix.
• Lower correlations (Pearson R < 0.9) were observed in specific genomic regions.
• Regions with lower agreement were frequently associated with ENCODE blacklist regions or gene clusters.

Conclusions:

• Updated LAI methods largely agree with previous approaches in admixed populations.
• Genomic sequence context, particularly ENCODE blacklist regions, contributes to lower inference agreement.
• Inference stability is more influenced by genomic region characteristics than by LAI software or genotyping technology.