Open SangGold opened 4 months ago
h2 estimates from ldsc can be negative, there is no error there; if you actually got an error, please report the precise error message Then, either the trait you're looking at has a very very small heritability, or the matching between LD scores and chi square stats is incorrect.
Any update on this?
when i run LDpred2 with gwas of biobank Japan for parkinson disease and target genotypes data is 504 samples of EAS in 1KGP, it raise an error: ldsc h2 estimate value is negative (h2 = -0.000936). I don't know why and how to fix it, could you help me resolve this problem?! Thank you so much!