The global test detected horizontal pleiotropy, but the outlier test detected no outlier.

[yitangsun]

Thank you for providing the great package.

I found that the global test detected horizontal pleiotropy, but the outlier test detected no outlier. So how do we interpret and deal with that?

Thank you!

[littleteacupscu]

I also have some issue

[littleteacupscu]

I found an answer from Marie: Hi again, it is possible that MR-PRESSO detects horizontal pleiotropy (significant P-value for the global test) but is not able to pinpoint any outlier (no significant P-value for any of the outlier test). In that case, no outlier-corrected estimate can be computed. Best, Marie

[marieverbanck]

Hi @yitangsun, @littleteacupscu, Indeed thanks @littleteacupscu please refer to issue #9. Unfortunately there is not much you can do. You could try and rerun it with a different seed just in case. Best, Marie

[steffzz04]

the same issue

[steffzz04]

the same issue

But I rerun the same code, the outcome become reasonable