search

brielin / Popcorn

Software for estimating correlation of trait effect sizes across populations
38 stars 15 forks source link

Inconsistent test results #16

Closed FEIFEI502 closed 3 years ago

FEIFEI502 commented 3 years ago

Dear Popcorn group,

Thanks for the fantastic software. But when I test the data provided in the test file, the pg is 0.480(se 0.166) whereas 0.302(se 0.053) in your README file. Since I used the version 1.0, is it any update about the method?

Thanks in advance, Fei-Fei

harmjanwestra commented 3 years ago

I'm getting the same values. Also I'm getting the following runtime warnings, which may be causing this issue?

[..]/lib/python2.7/site-packages/scipy/stats/_distn_infrastructure.py:879: RuntimeWarning: invalid value encountered in greater return (self.a < x) & (x < self.b) [..]/lib/python2.7/site-packages/scipy/stats/_distn_infrastructure.py:879: RuntimeWarning: invalid value encountered in less return (self.a < x) & (x < self.b) [..]/lib/python2.7/site-packages/scipy/stats/_distn_infrastructure.py:1738: RuntimeWarning: invalid value encountered in greater_equal cond2 = (x >= self.b) & cond0

brielin commented 3 years ago

Hi, sorry I missed this the first time. You will get one value with --use_mle and the other without it. I probably forgot to change the readme when I changed the default. Let me know if this doesn't work. The warnings should be fine although I should update the code at some point to get rid of them.