TrustPCA is an advanced tool for Principal Component Analysis (PCA) of ancient human genomic data, providing an estimation of projection uncertainty of genotype individuals.
Using a modern West Eurasian reference PC space, TrustPCA enables the projection of ancient individuals, similar to SmartPCA. However, TrustPCA goes further by estimating projection uncertainties based on missing loci in the provided samples. These uncertainties are visualized as ellipses on PCA plots, representing different confidence levels for sample placement.
All outputs, including statistical summaries and visualizations, can be exported as PDFs for seamless integration into research workflows.
Visit PTMVision at https://trustpca-tuevis.cs.uni-tuebingen.de/
GitHub: https://github.com/Integrative-Transcriptomics/trustpca