Table 1.
Associations of ANGPTL4 loss-of- function variants and type 2 diabetes
| ANGPTL4 LoF carriers | Total | Frequency | ||||
|---|---|---|---|---|---|---|
| Study | Cases | Controls | Cases | Controls | Cases | Controls |
| DiscovEHR | 22 | 85 | 12,969 | 36,217 | 0.0008 | 0.0012 |
| DiscovEHR 30K replication | 5 | 47 | 3,456 | 22,372 | 0.0007 | 0.0011 |
| UPenn | 3 | 13 | 734 | 4,066 | 0.0020 | 0.0016 |
| Duke | 2 | 10 | 1,630 | 4,903 | 0.0006 | 0.0010 |
| TAICHI | 0 | 3 | 4,392 | 4,699 | 0.0000 | 0.0003 |
| DHS-EA | 0 | 5 | 104 | 1,255 | 0.0000 | 0.0020 |
| DHS-AA | 0 | 1 | 357 | 2,028 | 0.0000 | 0.0002 |
| T2D-Genes/GoT2D/DIAGRAM | 14 | 19 | 8,373 | 8,466 | 0.0008 | 0.0011 |
| Total | 46 | 183 | 32,015 | 84,006 | 0.0007 | 0.0011 |
AA African American, CI confidence interval, Duke Duke CATHGEN cohort, EA European American, TD2-Genes/GoT2D/DIAGRAM combined analysis of T2D-GENES, GoT2D and DIAGRAM studies, pLoF predicted loss-of-function variant, Penn University of Pennsylvania Medicine Biobank, TAICHI TAIwan MetaboCHIp consortium
The overall odds ratio for ANGPTL4 pLoFs and type 2 diabetes risk, using a two-sided exact conditional test, was 0.71 (95% CI 0.49–0.99, p = 0.041)