search

n-mounier / MRlap

R package to perform two-sample Mendelian Randomisation (MR) analyses using (potentially) overlapping samples
GNU General Public License v2.0
49 stars 11 forks source link

rg value greater than 1 #17

Open TailinWang1 opened 7 months ago

TailinWang1 commented 7 months ago

Hi. When I use the function "MRlap", there is always a problem: the result comes out with an rg value greater than 1 every time. is there anyone like me? Can you tell me why?Thanks a lot! image

lcpilling commented 7 months ago

Hi. This estimate comes from LDSC and has been discussed before on their GitHub. https://github.com/bulik/ldsc/issues/89

The short answer is that the rG estimate is not bound by [-1,1] -- and in your case the SE is very large (6.4!) so the estimate itself is not particularly meaningful

TailinWang1 commented 7 months ago

Thank you very much!I get it.

王泰麟 @.***

 

------------------ 原始邮件 ------------------ 发件人: "n-mounier/MRlap" @.>; 发送时间: 2024年2月19日(星期一) 晚上6:54 @.>; @.**@.>; 主题: Re: [n-mounier/MRlap] rg value greater than 1 (Issue #17)

Hi. This estimate comes from LDSC and has been discussed before on their GitHub. bulik/ldsc#89

The short answer is that the rG estimate is not bound by [-1,1] -- and in your case the SE is very large (6.4!) so the estimate itself is not particularly meaningful

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

MenglinLu commented 7 months ago

你好，请问是怎么直接用MRlap得到的rg,rg_se和rg_p呢，可以给个代码示例看看吗，看教程好像只能输出rg，谢谢