I think editing of the genes would have the same effect that medicine does, medicine basically tries to silence those genes but you have to take it everyday. With gene editing it would probably reverse hairloss / maintain depending on how many genes that relate to baldness you decide to eliminate. Are all the 250+ genes you guys mentioned in every single bald/balding guy? If you could compare the genes with those that bald young and those that bald late I'm sure you can see which genes are more prominent for balding and tackle those.Right, but those are only the SNPs that reached genome-wide significance in that particular study. Boyle, Yang, and Pritchard estimate that ~100,000 SNPs contribute to complex traits, but that most have tiny effects and escape detection.
So there could be loads of SNPs with odds ratios like 1.002 contributing to A.G.A heritability, but Hagenaars et al. didn't have the power to go below ~1.03 with "only" 52,000 people.
In terms of p-value that's always true (at least in Europeans, since the risk allele for the top hit near AR is fixed in East Asians). In terms of effect size, that's usually true as well, but the largest effect sizes in Hagenaars et al.'s dataset were in rare variants around RSPO2 on Chromosome 8, with ORs up to 6.12 reaching genome-wide significance.
(I added the p < 5e-8?, |Beta|, and OR columns into the summary data)
View attachment 66383
The region around AR on the X chromosome goes up to 1.79.
View attachment 66384
Variants around the SRD5A2 (5-alpha reductase type II) gene on Chr2 aren't far behind AR.
View attachment 66385
But the comparison between the AR and RSPO2 or SRD5A2 regions isn't fair, since they didn't have imputed data for the X chromosome. Maybe AR would win in OR too with imputed data...And those rare variants must have huge error bars as well.
What would be the advantage in doing that instead of silencing AR totally in scalp dermal papilla cells?
And another question: Do you think individual genome sequencing and editing might help with reversal and not just maintenance?