Figure 4

Sample sizes (numbers of cases and controls, y -axis) for power of 80%. The power loss due to increasing numbers M of unassociated SNPs may be offset by increased sample sizes. Figure 4 shows the numbers of cases and controls each (y-axis) that are necessary to maintain power of 80% in the presence of M of unassociated SNPs (x-axis). Parameter values are as in Figure 3. Pattern approach (solid red line) and single-locus approach (broken blue line) furnish virtually identical results.