Open Conxz opened 4 years ago
Thanks for your comment. We correct for multiple testing by adapting the significance threshold based on the number of tests performed (Number of genes across tissues). I see that our wording in the script was indeed confusing.
From: Xiangzhen Kong notifications@github.com Sent: Tuesday, 28 January 2020 11:51 PM To: eskederks/eMAGMA-tutorial eMAGMA-tutorial@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: [eskederks/eMAGMA-tutorial] --adap-permp not for multiple testing correction (#4)
--adap-permp should be for obtaining permutation-based empirical gene p-values. It's not for multiple testing correction.
You can view, comment on, or merge this pull request online at:
https://github.com/eskederks/eMAGMA-tutorial/pull/4
Commit Summary
File Changes
Patch Links:
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/eskederks/eMAGMA-tutorial/pull/4?email_source=notifications&email_token=AKTD7YT5AUHSYYIALTRNPPLRAAZ45A5CNFSM4KMSYFV2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IJHJE6Q, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AKTD7YTBLL2ALIGE3735XQDRAAZ45ANCNFSM4KMSYFVQ.
Great. That's clear now. I raised this because a student asked me about this issue after reading this eMAGMA tutorial. Permutation-based testing and permutation-based correction always make people confused.
Yes, I agree! I accepted your change, thanks again
From: Xiangzhen Kong notifications@github.com Sent: Wednesday, 29 January 2020 10:46 PM To: eskederks/eMAGMA-tutorial eMAGMA-tutorial@noreply.github.com Cc: Eske Derks Eske.Derks@qimrberghofer.edu.au; Comment comment@noreply.github.com Subject: Re: [eskederks/eMAGMA-tutorial] --adap-permp not for multiple testing correction (#4)
Great. That's clear now. I raised this because a student asked me about this issue after reading this eMAGMA tutorial. Permutation-based testing and permutation-based correction always make people confused.
— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/eskederks/eMAGMA-tutorial/pull/4?email_source=notifications&email_token=AKTD7YWRS46NWFBV5WJJG2TRAF3AVA5CNFSM4KMSYFV2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKHCBGI#issuecomment-579739801, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AKTD7YT4A4WHKV2RPTYEOHLRAF3AVANCNFSM4KMSYFVQ.
--adap-permp should be for obtaining permutation-based empirical gene p-values. It's not for multiple testing correction.