Closed ANRudrapatna closed 7 months ago
emiraldi
commented
7 months ago I agree with Xiaoting -- provided peak centric is working as expected (predictions limited to a set of .bed intervals), let's keep it. As I recall, there might have been some issue that peak centric sometimes outputted predictions outside of the .bed intervals, due to overhang -- but this should be easy to fix with a bedtools intersection clean-up step at the end.
ANRudrapatna
commented
7 months ago Hello everybody:
This is likely an edit I proposed for v1.0.5, where prediction restricted to an input set of intervals did not work (see some of my previous emails comparing whole-genome vs. "peak-centric" mode for maxATAC predict). It looks like this now works in v1.0.6, so we can ignore these edits!
Please let me know if you have any questions!
Akshata Rudrapatna
From: Emily Miraldi @.> Sent: Tuesday, January 23, 2024 2:03:08 PM To: MiraldiLab/maxATAC @.> Cc: Rudrapatna, Akshata (rudrapan) @.>; Author @.> Subject: Re: [MiraldiLab/maxATAC] Update predict.md (PR #124)
External Email: Use Caution
I agree with Xiaoting -- provided peak centric is working as expected (predictions limited to a set of .bed intervals), let's keep it. As I recall, there might have been some issue that peak centric sometimes outputted predictions outside of the .bed intervals, due to overhang -- but this should be easy to fix with a bedtools intersection clean-up step at the end.
— Reply to this email directly, view it on GitHubhttps://github.com/MiraldiLab/maxATAC/pull/124#issuecomment-1906738752, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BBKU2MJOX5SKLNZGBEGX5HLYQACOZAVCNFSM6AAAAAA646PLICVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMBWG4ZTQNZVGI. You are receiving this because you authored the thread.Message ID: @.***>
Removed "peaks" as an option for maxATAC, given potential issues with the peak-centric mode in maxATAC.