| PMC full text: | Published online 2012 Sep 25. doi: 10.1371/journal.pone.0044507
|
Table 1
Frequency and relative effect size of haplotype between rs11878604 (SNP) and rs8102683 (CNP).
| Haplotype¶ | GWAS set | Replication set | |||||
| rs11878604 (SNP) | rs8102683 (CNP) | Haplotype Freq.* (%) | Relative effect size* (s.e.) | P * | Haplotype Freq.* (%) | Relative effect size* (s.e.) | P * |
| T | 1 copy | 41.41 | reference | 41.8 | reference | ||
| T | 0 copy | 19.22 | −4.00 (0.290) | 3.8×10−43 | 18.6 | −4.30(0.384) | 8.3×10−29 |
| C | 1 copy | 38.14 | −2.69 (0.237) | 9.4×10−30 | 38.4 | −2.63(0.309) | 2.3×10−17 |
| C | 0 copy | 0.76 | −5.05 (1.548) | 1.1×10−3 | 0.63 | −3.84(2.200) | 0.081 |
| C | 2 copies | 0.23 | −3.07 (2.900) | 0.30 | 0.24 | −2.62(3.612) | 0.47 |
| T | 2 copies | 0.28 | −0.29 (2.413) | 0.90 | 0.31 | −0.414(2.879) | 0.89 |
| Explained variance† | 1.83% (ANOVA F-test P =9.5×10−52) | 2.23% (ANOVA F-test P=2.8×10−30) | |||||
¶Haplotypes consisting of alleles at rs11878604 [T/C] and haplotypic copy numbers at rs8102683 [0/1/2].
*Haplotype frequencies and relative effect sizes are jointly evaluated using a haplotype-specific linear regression model as described previously [27]. All model parameters were estimated using a standard EM (expectation-maximization) algorithm [28].
†Variance explained by these six haplotypes and its significance was assessed in the standard ANOVA (analysis of variance) framework.