Table 7.
Class | Crude (Fisher’s exact) | Adjusted |
---|---|---|
Heterogeneity across racial/ethnic groups by sequence variant class |
|
|
Racial/ethnic group: T + SJV |
0.11 |
NA¥ |
Racial/ethnic group: any rMS |
0.35 |
NA¥ |
subset: non-key domain rMS*: |
0.40 |
NA¥ |
subset: key domain rMS*: |
0.003 |
NA¥ |
sub-subset: key domain rMS*, <C65 |
0.96 |
NA¥ |
sub-subset: key domain rMS*, C65 |
<1x10−5 |
NA¥ |
Heterogeneity across genes, by likely pathogenic sequence variant class |
|
|
MRN: T + SJV |
0.43 |
NC† |
MRN: key domain rMS* |
1.00 |
0.68 |
MRN: rare T + SJV and key domain rMS* | 0.53 | 0.30 |
¥We have adjusted most P values for racial/ethnic group and study center. But a test for heterogeneity across racial ethnic groups cannot be adjusted for racial/ethnic group; moreover, as most of the subjects of East Asian, Latina, and Recent African Ancestry were obtained from one study center (the Cancer Prevention Institute of California), we cannot adjust these tests on that variable, either. †Adjusted P value could not be calculated because some cells contained 0. *The key functional domains are defined in Figure 1. The set of variants includes rare missense substitutions with A-GVGD scores > C0, and final exon nonsense and frameshift variants if they also fall in a key functional domain. T + SJV, truncating and splice junction variants; rMS, rare missense substitution; MRN, MRE11-RAD50-NBN complex.