IPCAPS: an R package for iterative pruning to capture population structure

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Resolving population genetic structure is complex, particularly for closely related or geographically isolated groups.
• Principal Component Analysis (PCA) and single nucleotide polymorphisms (SNPs) are common for analyzing shared genetic ancestry.
• Enhancements are needed for accurately detecting fine-scale population structure.

Purpose of the Study:

• To introduce IPCAPS, an R package designed for resolving fine-scale population genetic structure.
• To provide a robust tool for analyzing genetic variation and population stratification.

Main Methods:

• IPCAPS utilizes the iterative pruning Principal Component Analysis (ipPCA) framework.
• The method systematically assigns individuals into genetically similar subgroups.
• Outlier detection and elimination are performed iteratively to prevent misclassification.

Main Results:

• The IPCAPS package effectively resolves fine-scale population structure using SNP data.
• The iterative pruning approach enhances accuracy by identifying and removing outliers.
• The tool demonstrates systematic assignment of individuals to genetically homogeneous subgroups.

Conclusions:

• IPCAPS accommodates various measurement scales, including gene expression and methylation data.
• The package is applicable to patient sub-phenotyping, aiding in clinical research.
• IPCAPS is freely available as an R package from http://bio3.giga.ulg.ac.be/ipcaps.