Registration for a live webinar on 'Phasing Out Animal Testing: A New Era for Pharmaceutical Innovation' is now open.
See webinar details
This is a limited length demo talk; you may
login or
review methods of
obtaining more access.
- Special Designs
-
46 min
1. Family-based association tests: introduction- Dr. Christoph Lange
-
29 min
-
37 min
3. Two-stage genome-wide association designs- Dr. Andrew Skol
- Special Populations
-
40 min
4. Admixture mapping- Dr. Paul McKeigue
- Whole-Genome Studies: Types of Data
-
34 min
5. QTL association mapping- Prof. Bruce Weir
-
50 min
6. Adjusting for population structure in genetic association studies- Prof. David Balding
- Whole-Genome Studies 2: Methods
-
47 min
- Archived Lectures *These may not cover the latest advances in the field
-
49 min
8. Family-based association tests: FBATs for various data types- Dr. Christoph Lange
-
44 min
-
54 min
10. Use of isolated populations: pitfalls and potential- Dr. Leena Peltonen
-
40 min
-
37 min
12. Detecting multiple associations in genome-wide studies- Prof. Frank Dudbridge
-
30 min
13. Genome scanning by composite likelihood- Dr. Andrew Collins
-
41 min
14. Multistage sampling for genetic studies- Prof. Robert Elston
-
39 min
-
43 min
16. A survey of multi-locus methods for analysis of large SNP datasets- Mr. A. Geert Heidema
-
34 min
17. Sequential analysis methods in genomic scans- Prof. Michael Province
Printable Handouts
Navigable Slide Index
- Introduction
- Background (1)
- Background (2)
- Objectives of stage 1 analysis
- Typical strategy
- Problem
- Case-control study: single SNP power
- Main effects versus joint and interaction effects
- Epistasis with no main effects
- Epistasis with main effect
- Epistasis main effect
- Multiplicative models with heterogeneity (1)
- Multiplicative models with heterogeneity (2)
- Multiplicative models - two locus example
- Two locus example - 2+ independent systems
- Formal definition of the model
- Example genetic models
- Random forest analysis
- Random forest (1)
- Random forest (2)
- Random forest variable importance
- Random forest importance (1)
- Random forest importance (2)
- Simulation study
- Genetic models
- Genetic models for simulation
- Simulation study plan (1)
- Simulation study plan (2)
- Rank of risk SNPs: N = 100 SNPs (4 risk SNPs)
- Rank of risk SNPs: N = 1000 SNPs (4 risk SNPs)
- Rank of risk SNPs: N vs. proportion (1)
- Rank of risk SNPs: N vs. proportion (2)
- Simulation study plan (3)
- Rank of risk SNPs: H8M4 (1)
- Rank of risk SNPs: H8M4 (2)
- Conclusions
- Effects of linkage disequilibrium (1)
- Effects of linkage disequilibrium: an illustration
- Effects of linkage disequilibrium (2)
- Effects of linkage disequilibrium (3)
- Genome-wide association analysis (1)
- Genome-wide association analysis (2)
- GAW 15 simulated 10K data (1)
- GAW 15 simulated 10K data (2)
- GAW 15 analysis
- Variable selection
- GAW 15 analysis: prediction error
- GAW 15 analysis results: power
- Computational issues
- Summary (1)
- Summary (2)
- Random Forest software
- References
- Acknowledgments
Topics Covered
- Screening association studies with a two-stage approach
- Multiplicative models with heterogeneity
- Random forest analysis
- Random forest variable importance
- Power
- Effect of linkage disequilibrium
- Genome-wide association studies
- Computational issues
Talk Citation
Lunetta, K. (2007, October 1). Screening large-scale association study data: exploiting multi-gene interactions with random forests [Video file]. In The Biomedical & Life Sciences Collection, Henry Stewart Talks. Retrieved March 28, 2025, from https://doi.org/10.69645/ETKX5510.Export Citation (RIS)
Publication History
Financial Disclosures
- Dr. Kathryn Lunetta has not informed HSTalks of any commercial/financial relationship that it is appropriate to disclose.
Screening large-scale association study data: exploiting multi-gene interactions with random forests
A selection of talks on Methods
Hide