Results: 4

1.
Fig. 4

Fig. 4. From: Including Additional Controls from Public Databases Improves the Power of a Genome-Wide Association Study.

Genomic inflation factor λ versus number of principal components used for correction. There are 21 principal components that are significant using Tracy-Widom test statistics when the control:case ratio is 20: 1.

Semanti Mukherjee, et al. Hum Hered. 2011 September;72(1):21-34.
2.
Fig. 1

Fig. 1. From: Including Additional Controls from Public Databases Improves the Power of a Genome-Wide Association Study.

Analytical power of GWAS with additional controls. All power calculations assume an additive model and significance level of α = 1 × 10-7. The power computed using GRR of 1.2, 1.4, and 1.6 and DAF of 0.1, 0.2, and 0.4 is plotted.

Semanti Mukherjee, et al. Hum Hered. 2011 September;72(1):21-34.
3.
Fig. 3

Fig. 3. From: Including Additional Controls from Public Databases Improves the Power of a Genome-Wide Association Study.

Quantile-quantile plot of GWAS of pancreatic cancer cases with additional controls. At a 20: 1 control:case ratio, this plot compares the association statistics without any population stratification correction (green), after correction with principal components analysis (red), or with both PCA and removal of SNPs that show differences between the MSKCC controls and additional controls (blue). The black line shows the expected result under the null hypothesis of no association.

Semanti Mukherjee, et al. Hum Hered. 2011 September;72(1):21-34.
4.
Fig. 2

Fig. 2. From: Including Additional Controls from Public Databases Improves the Power of a Genome-Wide Association Study.

Population substructure of the MSKCC pancreatic cancer study and additional controls. Principal components were computed for the MSKCC and additional controls samples combined, and plotted separately. a, c Principal components of the 263 cases and 202 controls from the MSKCC (New York) pancreatic cancer study. The first principal component is plotted against the second (a) or third (c). Individuals in red self-identified as Ashkenazi Jewish in the study questionnaire. b, d Principal components of the additional controls from dbGaP. The first principal component is plotted against the second (b) and third (d) principal components.

Semanti Mukherjee, et al. Hum Hered. 2011 September;72(1):21-34.

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Write to the Help Desk