Genome scans for Q1 and Q2 on general population replicates using Loki

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Genome scans are crucial for identifying genes associated with complex traits.
• Markov Chain Monte Carlo (MCMC) methods offer a powerful computational approach for linkage analysis.

Purpose of the Study:

• To evaluate the effectiveness of the Loki MCMC linkage package for genome scans.
• To assess the impact of increased data and analytical complexity on quantitative trait loci (QTL) detection.
• To optimize MCMC sampling strategies for efficient genome-wide linkage analysis.

Main Methods:

• Genome scans were performed using the Loki MCMC linkage package on replicate datasets.
• Analyses varied in marker data availability, inclusion of polygenic effects, and chromosome-wise versus joint analysis.
• Sampler convergence was assessed by repeating analyses with different random number seeds.

Main Results:

• The initial genome scan successfully detected and localized major genes (MG1 for Q1, MG3 for Q2).
• Incorporating additional data (full marker data, polygenic effects, joint chromosome analysis) enhanced linkage signal quality and reduced false positives.
• Increased data did not lead to the detection of QTLs missed in the initial, less complex analysis.
• MCMC sampler convergence was confirmed through repeated analyses.

Conclusions:

• The Loki package is effective for MCMC-based genome scans.
• A two-stage approach—initial single-chromosome scans followed by multi-chromosome analysis of promising regions—is an efficient strategy for this dataset.
• While additional data improves analysis robustness, it may not rescue missed QTLs from initial scans.