Table 1. Direct identification of the gene for a Mendelian disorder by exome resequencing.
Each cell indicates the number of genes with nonsynonymous (NS) variants, splice acceptor and donor site mutations (SS), and coding indels (I). Filtering by either requiring the presence of NS/SS/I variants in siblings (kindred 1 (A+B)), multiple unrelated individuals (columns), or by excluding annotated variants (rows) identifies 26 and 8 candidate genes under a dominant model and only a single candidate gene, DHODH, under a recessive model (light grey cells). Exclusion of mutations predicted to be benign using PolyPhen (row 5) increases sensitivity under a dominant model but excludes DHODH under a recessive model because a variant in kindred 1 is predicted to be benign. A single candidate gene is identified in kindred 1 under a recessive model (dark grey cell) but this candidate is excluded in comparisons with unrelated cases of Miller syndrome. Mutations in this candidate, DNAH5, were found to cause a primary ciliary dyskinesia in kindred 1. The asterisk indicates that a second gene, CDC27, was also identified as a candidate gene but on manual review it appears to be multiple copies of a processed pseudogene that cause a false positive.
| Filter | Kindred 1-A | Kindred 1-B | Kindred1 (A + B) | Kindreds 1+2 | Kindred 1+2+3 | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Dominant | Recessive | Dominant | Recessive | Dominant | Recessive | Dominant | Recessive | Dominant | Recessive | |
| NS/SS/I | 4670 | 2863 | 4687 | 2859 | 3940 | 2362 | 3099 | 1810 | 2654 | 1525 |
| … not in dbSNP129 | 641 | 102 | 647 | 114 | 369 | 53 | 105 | 25 | 63 | 21 |
| … notin HapMap 8 | 898 | 123 | 923 | 128 | 506 | 46 | 117 | 7 | 38 | 4 |
| … not in either | 456 | 31 | 464 | 33 | 228 | 9 | 26 | 1* | 8 | 1* |
| … AND predicted damaging | 204 | 6 | 204 | 12 | 83 | 1 | 5 | 0 | 2 | 0 |