Numerous Divide By Zero Warnings

[DarioS]

I see thousands of warnings about dividing by zero. Should I, as an end-user. be concerned and do anything differently?

diffacto.py:171: RuntimeWarning: divide by zero encountered in double_scalars
diffacto.py:208: RuntimeWarning: invalid value encountered in true_divide

It also happens for the HBY20 Mix example data set. Sometimes, ss_resid is 0. Please document this phenomenon for users to know more about it. I also frequently see -inf in the S/N column. Should the columns of the results table also be documented?

[percolator]

Hi Dario, I think this relates to some type of missing values. Would you be able to share data, so that we can recreate the error. Yours --Lukas

…

On Wed, Apr 21, 2021 at 6:00 AM Dario Strbenac ***@***.***> wrote: I see thousands of warnings about dividing by zero. Should I, as an end-user. be concerned and do anything differently? diffacto.py:171: RuntimeWarning: divide by zero encountered in double_scalars diffacto.py:208: RuntimeWarning: invalid value encountered in true_divide It also happens for the HBY20 Mix example data set. Sometimes, ss_resid is 0. Please document this phenomenon for users to know more about it. I also frequently see -inf in the S/N column. Should the columns of the results table also be documented? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#19>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAXKADQ5ACVIMCYLZ7TKPDTJZEVLANCNFSM43JQRTTA> .

[DarioS]

It can be triggered by running Example-2 (HBY20) in the examples directory of this repository using the example command shown.

[rcpoulos]

I would also like to query the -inf values often returned in the S/N column. Should these proteins be treated as below the S/N threshold (eg -20) and therefore be discarded?

[percolator]

Yes, proteins with S/N of -inf are indeed below any threshold and should be treated with caution. --Lukas

…

On Fri, Aug 6, 2021 at 6:04 AM Rebecca Poulos ***@***.***> wrote: I would also like to query the -inf values often returned in the S/N column. Should these proteins be treated as below the S/N threshold (eg -20) and therefore be discarded? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#19 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAXKAG4SCBISTGVWWZO3VLT3NNMBANCNFSM43JQRTTA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email> .