The use of plasmodes as a supplement to simulations: A simple example evaluating individual admixture estimation...

Area of Science:

• Statistical genetics
• Computational biology
• Bioinformatics

Background:

• Computer simulations are crucial for statistical methodology research but can lack realism.
• Admixture estimation, determining genomic ancestry proportions, is challenging with complex population structures.
• Plasmode datasets, derived from natural biological processes with known truths, offer a realistic alternative to simulations.

Purpose of the Study:

• To evaluate the effectiveness of admixture estimation methodologies using plasmode datasets.
• To demonstrate the utility of plasmodes in assessing statistical genetics methods.
• To compare the performance of different admixture estimation algorithms under varying population complexities.

Main Methods:

• Utilized mouse cross data as plasmode datasets where ancestry proportions are known.
• Applied established admixture estimation methodologies (Structure, AdmixMap, FRAPPE).
• Compared estimated admixture proportions against known true proportions in simple and complex (three-population) plasmodes.

Main Results:

• All tested admixture estimation methods performed well on simple plasmode datasets.
• Method performance varied significantly when applied to a three-population plasmode dataset.
• Plasmodes effectively highlighted differences in methodology performance with increasing population complexity.

Conclusions:

• Plasmode datasets are valuable for evaluating the accuracy and reliability of statistical genetics methods.
• The performance of admixture estimation techniques is sensitive to population structure complexity.
• Plasmodes provide a robust framework for advancing statistical genetics methodology research.