Table 1. Summary statistics and counts of cis- and trans-eQTLs at thresholds by R2 or |β|.
| cis-eQTL(n = 3,804 by FDR <5%) | trans-eQTL(n = 165 by FWER <5%) | ||||||||
| All | mRNA | lincRNA | Other | All | mRNA | lincRNA | Other | ||
| #eQTLs-transcript pairs | 3,804 | 2,995 | 293 | 516 | 165 | 91 | 49 | 25 | |
| #unique eQTLs | 3,385 | 2,779 | 244 | 440 | 105 | 65 | 34 | 21 | |
| #unique transcripts (%) | 3,804 (12.5%) | 2,995 (15.1%) | 293 (4.8%) | 516 (11.6%) | 114 (0.4%) | 60 (0.3%) | 34 (0.6%) | 20 (0.4%) | |
| #unique genes (%) | 2,973 (16.9%) | 2,667 (17.3%) | 0 | 357 (10.9%) | 74 (0.4%) | 57 (0.4%) | 0 | 17 (0.5%) | |
| #unique transcripts without gene ID | 455 | 28 | 293 | 134 | 54 | 1 | 49 | 4 | |
| R 2 mean±SD | 0.19±0.15 | 0.19±0.15 | 0.20±0.16 | 0.21±0.18 | 0.27±0.12 | 0.27±0.12 | 0.29±0.13 | 0.23±0.07 | |
| R 2 median±IQR | 0.13±0.15 | 0.13±0.14 | 0.13±0.15 | 0.15±0.17 | 0.23±0.12 | 0.23±0.12 | 0.26±0.11 | 0.21±0.09 | |
| |β| mean±SD | 0.33±0.33 | 0.32±0.32 | 0.38±0.33 | 0.38±0.34 | 0.53±0.35 | 0.50±0.38 | 0.59±0.31 | 0.51±0.33 | |
| |β| median±IQR | 0.24±0.24 | 0.23±0.23 | 0.30±0.31 | 0.28±0.25 | 0.47±0.41 | 0.40±0.44 | 0.58±0.34 | 0.43±0.34 | |
| Stat. | Cutoff | ||||||||
| R 2 | 0.1 | 2,568 | 2,022 | 192 | 354 | 165 | 91 | 49 | 25 |
| 0.3 | 665 | 500 | 56 | 109 | 45 | 25 | 15 | 5 | |
| 0.5 | 245 | 171 | 22 | 52 | 10 | 5 | 5 | 0 | |
| 0.7 | 67 | 43 | 8 | 16 | 2 | 2 | 0 | 0 | |
| 0.9 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | |
| |β| | 0.3 | 1,440 | 1,053 | 146 | 241 | 118 | 60 | 40 | 18 |
| 0.9 | 155 | 122 | 11 | 22 | 23 | 13 | 6 | 4 | |
| 1.5 | 55 | 44 | 3 | 8 | 3 | 1 | 1 | 1 | |
| 2.1 | 26 | 19 | 2 | 5 | 1 | 1 | 0 | 0 | |
| 2.7 | 12 | 9 | 1 | 2 | 1 | 1 | 0 | 0 | |
FDR: false discovery rate; FWER: family-wise error rate; SD: standard deviation; IQR: inter-quartile range; R 2: proportion of phenotypic variances explained by genotypes; |β|: absolute value of coefficient of genotypes.
The sum of #unique eQTLs counted within RNA types is not necessarily equal to #unique eQTLs counted for all transcripts because the same eQTLs may be counted in more than one RNA types. The number of genes for All and each type do not match for a similar reason.