search

weizhouUMICH / SAIGE

GNU Lesser General Public License v3.0
187 stars 72 forks source link

The appropriate sample size for continuous and binary traits analysis in SAIGE-GENE+ #445

Open jxcao98 opened 6 months ago

jxcao98 commented 6 months ago

Hi,

Thanks for your great tool!

I noticed that SAIGE-GENE+ is perfectly suited for large-scale cohorts such as UKB. Now I wonder if it works for much smaller cohorts, like 300 cases vs. 300 controls.

Also, I received an error: "ERROR! The first variance component parameter estimate is 0" when I ran Step 1 on about 1,000 samples. Does this have anything to do with the small sample size?

Is there any guidance for the recommended sample size for the application of SAIGE-GENE+ to continuous and binary traits?

Thanks in advance!

Best regards, Jixin