Skip to main content
. Author manuscript; available in PMC: 2024 Aug 27.
Published in final edited form as: Nat Genet. 2023 Nov 30;55(12):2235–2242. doi: 10.1038/s41588-023-01562-0

Extended Data Fig. 5 |. Pseudo-R2 for noncoding regions.

Extended Data Fig. 5 |

Pseudo-R2 is calculated for noncoding regions for two datasets: gnomAD v3 and UK Biobank. Since Roulette was trained on noncoding variants from the gnomAD v3, it is expected that Roulette performs better for noncoding variants than synonymous variants. De novo sequencing and UK Biobank population sequencing is an independent dataset from trained data.