Hi everyone,
I’m working on an evolutionary analysis using the gnomAD v4.1.0 dataset. I’ve noticed that for some genes, specifically ACE2 (ENSG00000159640), the constraint metrics (Missense Z-score, pLI, o/e) are missing from the browser and the summary tables, although they were available in v2.1.1.
Does anyone know if:
-
There is a specific filtering criteria (e.g., coverage or low complexity) that excluded ACE2 from the v4 constraint model?
-
There’s a known ETA for the release of these metrics for the remaining genes in v4?
If anyone has calculated “in-house” Z-scores for v4 or has a workaround for these missing values in ML features, I’d love to hear your thoughts.
Thanks in advance!
Hello,
As we detailed in this blog post, the constraint scores originally released for gnomAD v4.0 were missing for chromosomes X and Y. Since ACE2 is on chrX, it is currently missing a score and the suggestion was to use the v2-based scores.
We are planning to release v4.1-based constraint scores in the next month or so that will be available for chrX. Our apologies for the delay on getting these out to the community.
Thanks,
Kaitlin
Congratulations to the gnomAD team for the continued transparency and rigor in addressing constraint metric coverage, particularly for sex chromosomes. Accounting for X- and Y-linked genes has long been a structural challenge in population-scale constraint modeling, with real downstream implications for disease interpretation and target assessment. Clear communication around limitations, interim guidance, and timelines—such as the forthcoming v4.1-based chrX constraint scores—meaningfully advances the reliability and usability of these resources for both research and translational applications. It is encouraging to see steady progress toward more complete genomic representation, and I look forward to the impact this will have on variant interpretation, gene prioritization, and therapeutic discovery.