1. STANDARD Frequency of the putative susceptibility allele in cases is 0.3, and frequency in controls is 0.2. Assume case/control ratio of 1. Without using Internet, figure out
   1. What is the power to detect this allele at nominal P=0.05 in a sample of 30 cases?
   2. What sample size you need to detect it with power of 90% at nominal P=0.05?
   3. What sample size you need to detect it with power of 80% at genome-wide P=0.05, in a study assessing 500,000 SNPs?
2. STANDARD Read paper of Scott et al., Science, 2007. This study identifies novel T2D locus, IGF2BP2. You have a sample of 600 T2D cases and 1200 controls. Use genetic power calculator (http://pngu.mgh.harvard.edu/~purcell/gpc/) to answer following questions.
   1. What is the power to detect this polymorphism in your sample, when you do single test (candidate polymorphism study)?
   2. If you used your sample to do GWA analysis with 300K SNPs, what would be the power to detect this polymorphism at 5% GW-significance level?
   3. If you sample consisted of 600 trios (affected child + parents, suited for TDT test), what power would you have to detect this gene (also, as candidate polymorphism, and in a GWA study with 300K)?
   PS Did you pay attention to the fact that in the paper, you are supplied with Odds Ratio, but genetic power calculator asks you for genotype relative risks? Also, the paper gave you the frequency of allele in CONTROLS, while genetic power calculator asks you for population frequency.
3. In a population, average inbreeding coefficient is 1/16. Frequency of an allele is 0.1. Assuming that your data exactly follow the model, try to figure out what sample size you need to detect detect deviation from HWE due to inbreeding? Try at least to outline the solution.
4. MEDIUM to HIGH The PPARgamma risk allele has frequency of 0.7. Presence of each copy of this allele increases the risk of type 2 diabetes by 1.2, compared to risk in non-carriers (8%). Without using Internet, tell what is the sample size requred to detect this allele
   1. At alpha of 0.05 and power of 0.8 in a case-control candidate polymorphism study
   2. GWA case-control study with 500,000 SNPs, at GW-significant level (power 0.8)?
   3. Verify your results using genetic power calculator on Internet.