Printable Handouts

PDF

Navigable Slide Index

1. Introduction
2. Effect of population structure on association (1)
3. Effect of population structure on association (2)
4. Spurious signals of association
5. Genealogy at causal locus (haploid)
6. Genome-wide allele sharing due to pedigree
7. The cryptic substructure problem
8. Generation of spurious associations
9. Pre-2006 approaches to population structure
10. TDT: pros and cons
11. Genomic control (GC)
12. Test statistics after genomic control adjustment
13. Use of GC adjustment
14. GC: pros and cons
15. Structured association
16. Problems with SA methods
17. Detecting individuals with unusual ancestry
18. New approaches in 2006: logistic regression
19. Simulation study: statistical methods
20. Power of different statistical methods
21. Simulation with more null markers
22. Simulation involving ascertainment bias
23. New approaches in 2006: Eigenstrat
24. Simulation results for two populations
25. Logistic regression with/without PC adjustment
26. First 2 PCs ability to capture population structure
27. Eigenstrat: pros and cons
28. New approaches, 2006: mixed model approach
29. Mixed regression model
30. Mixed model approach
31. Discussion
32. Conclusions
33. References

Topics Covered

• Effect of population structure in generating spurious associations
• Cause of problem is confounded by pedigree
• Pre-2006 solutions: design based (TDT), genomic control, structured association
• New solutions in 2006: null markers as regression covariates, principal components as regression covariates, linear mixed models using pairwise kinships

Links

Series:

• Genetic Epidemiology II

Categories:

• Genetics & Epigenetics
• Methods

Talk Citation

Balding, D. (2007, October 1). Adjusting for population structure in genetic association studies [Video file]. In The Biomedical & Life Sciences Collection, Henry Stewart Talks. Retrieved April 19, 2024, from https://hstalks.com/bs/609/.
Export Citation (RIS)

Publication History

• Published on October 1, 2007

Financial Disclosures

• Prof. David Balding has not informed HSTalks of any commercial/financial relationship that it is appropriate to disclose.