Happy to introduce an ancestry estimation submodule in the hapla framework. If you have phased genotype data, unsupervised ancestry estimation has never been faster or more accurate! 🦾

Preprint: www.biorxiv.org/content/10.1...
Code: github.com/Rosemeis/hapla

We further find that reference-based PRSs computed from large-scale GWAS summary statistics on depression and schizophrenia are highly correlated with estimated principal components in the iPSYCH cohort, thus, indicating residual confounding due to population structure in the GWASs.

We show that we can utilize inferred haplotype clusters to compute more accurate heritability estimates and reference-free polygenic scores compared to standard genotype-based approaches in extensive simulations and the iPSYCH cohort for depression and schizophrenia disorders.

Delighted that our fastmixture manuscript has now been peer-reviewed and recommended in PCI Evol Biol, now 30x faster than ADMIXTURE! #openscience
doi.org/10.24072/pci...

BioRxiv: doi.org/10.1101/2024...
GitHub: github.com/Rosemeis/fas...