Inference and Analysis of Population Structure Using Genetic Data and Network Theory

Area of Science:

• Population Genetics
• Network Theory
• Computational Biology

Background:

• Genetic clustering is vital for understanding population structure.
• Existing methods (model-based, distance-based) have limitations like statistical rigor issues or unmet prior assumptions.

Purpose of the Study:

• To present a novel, model-free, distance-based approach for population structure inference using network theory.
• To provide a statistically rigorous method for analyzing genetic data without prior assumptions.

Main Methods:

• Constructing a genetic similarity network from individual genetic data.
• Applying community-detection algorithms to partition the network into subpopulations.
• Utilizing permutation tests to assess partition significance (modularity) and a new Strength of Association (SA) measure.

Main Results:

• Demonstrated applicability using human genetic data and simulations.
• Identified population structure, gene flow patterns, and potential hybrid individuals.
• The network approach proved computationally efficient and model-free.

Conclusions:

• The network theory-based method offers a robust, assumption-free alternative for population structure analysis.
• The approach enhances understanding of genetic relationships and population dynamics.
• The NetStruct software implements this novel methodology.