Family-Based Association Studies for Next-Generation Sequencing Yun Zhu, Momiao Xiong  The American Journal of Human Genetics  Volume 90, Issue 6, Pages 1028-1045 (June 2012) DOI: 10.1016/j.ajhg.2012.04.022 Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 1 Power Curves of the Family-Based CMC and Single-Marker Statistics as a Function of Sample Size (A) The power curves of the family-based CMC (variants with frequencies ≤ 0.005 were collapsed) statistic as a function of the total number of individuals at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, 20% of the risk variants, and a baseline penetrance of 0.01. (B) The power curves of the corrected single-marker statistic as a function of the total number of individuals at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, 20% of the risk variants, and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 2 Power Curves of the Family-Based FPCA and SFPCA Statistics as a Function of Sample Size (A) The power curves of the family-based FPCA statistic as a function of the total number of individuals at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, 20% of the risk variants and a baseline penetrance of 0.01. (B) The power curves of the family-based SFPCA statistic as a function of the total number of individuals at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, 20% of the risk variants, and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 3 Power of Family-Based CMC and Single-Marker Statistics as a Function of the Proportion of Risk Variants (A) The power curves of the family-based CMC (variants with frequencies ≤ 0.005 were collapsed) statistic as a function of the proportion of risk variants at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, a total of 1,800 sampled individuals, and a baseline penetrance of 0.01. (B) The power curves of the corrected single-marker statistic as a function of the proportion of rare variants at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, a total of 1,800 sampled individuals, and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 4 Power Curves of Family-Based FPCA and SFPCA Statistics as a Function of the Proportion of Risk Variants (A) The power curves of the family-based FPCA statistic as a function of the proportion of risk variants at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, a total of 1,800 sampled individuals, and a baseline penetrance of 0.01. (B) The power curves of the family-based SFPCA statistic as a function of the proportion of rare variants at the significance level α=0.05 in the test under seven settings: unrelated individuals in cases-controls study, nuclear family groups 1 and 2, sib-pair groups 1 and 2, and three-generational family groups 1 and 2, assuming the dominant model, a total of 1,800 sampled individuals, and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 5 Power of Tests as a Function of Sample Sizes for Group 1 Three-Generational Family and Sib-Pair (A) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for three-generational family group 1 as a function of the total number of individuals at the significance level α=0.05 under the dominant model, assuming 20% of risk variants and a baseline penetrance of 0.01. (B) The power curve of five family-based statistics: S FPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for sib-pair group 1 as a function of the total number of individuals at the significance level α=0.05 under the dominant model, assuming 20% of risk variants and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 6 Power of Tests as a Function of Sample Sizes for Group 1 Nuclear Family and Group 2 Three-Generational Family (A) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for nuclear family group 1 as a function of the total number of individuals at the significance level α=0.05 under the dominant model, assuming 20% of risk variants and a baseline penetrance of 0.01. (B) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for the three-generational family group 3 as a function of the number of sampled individuals at the significance level α=0.05 under the dominant model, assuming 20% of risk variants and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 7 Power of Tests as a Function of the Proportion of Risk Variants for Group 1 Three-Generational Family and Sib-Pair (A) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for the three-generational family group 1 as a function of the proportion of risk variants at the significance level α=0.05 under the dominant model, assuming a total of 1,800 sampled individuals and a baseline penetrance of 0.01. (B) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for sib-pair group 1 as a function of the proportion of risk variants at the significance level α=0.05 under the dominant model, assuming a total of 1,800 sampled individuals and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 8 Power of Tests for the Proportion of Risk Variants for Group 1 Nuclear Family and Group 2 Three-Generational Family (A) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for nuclear family group 1 as a function of the proportion of risk variants at the significance level α=0.05 under the dominant model, assuming a total of 1,800 sampled individuals and a baseline penetrance of 0.01. (B) The power curve of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for three-generational family group 2 as a function of the ratio of the number of rare risk variants over the number of common risk variants at the significance level under the dominant model, assuming a total of 1,500 sampled individuals and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions

Figure 9 Power Comparison of Tests for Group 2 Three-Generational Family under Opposite Directions of Association The power comparison of five family-based statistics: SFPCA, FPCA, CMC, generalized T2, and individual χ2 statistic for three-generational family group 2 under opposite directions of association as a function of the total number of individuals at the significance level α=0.05 under the dominant model, assuming 20% of the risk variants in one direction of association, 10% of risk variants and 10% of protective variants in two opposite directions of association, and a baseline penetrance of 0.01. The American Journal of Human Genetics 2012 90, 1028-1045DOI: (10.1016/j.ajhg.2012.04.022) Copyright © 2012 The American Society of Human Genetics Terms and Conditions