Table 3.
Summary of exome-based sequence variants in the 2 twin pairs
Twin pair | Detected variantsa | Variants not in dbSNP | Discordant variants | Discordant coding region variants predicted to potentially result in pathogenicityb | Predicted pathogenic discordant variants in affected twin only, with data above thresholdc | Discordant variants confirmed by Sanger sequencing |
---|---|---|---|---|---|---|
1 | 79,525 | 19,976 | 1,070 | 99 | 1 | 0 |
2 | 125,630 | 34,423 | 1,787 | 124 | 0 | 0 |
The difference in the number of variants in the 2 different twin pairs largely reflects different methods of exome sequencing (SureSelect Human All Exon 38-Mb vs. 50-Mb kits).
Differences in the number of detected variants reflect greater coverage in the second twin pair analyzed.
Includes deletion-insertion variants, non-synonymous variants, splice-site variants, and nonsense variants, but excludes variants in 3′UTR, 5′UTR, intron variants, noncoding variants, as well as synonymous variants and variants in repeat regions.
Our threshold score was set such that we required a most probable genotype score ≥10 and a score/coverage ratio ≥0.5 (for details see Patients and Methods under the Variant Analysis section). The false-positive discordant variant found in twin pair 1, which was shown to be an artifact by Sanger sequencing, was later found as an artifact in multiple samples sequenced at the same facility [Biesecker et al., 2009].