getWaveforms Bug: Spikes near start or end of recording

[nmtimme]

Hi,

Thanks for making this great resource!

I found a bug in the getWaveForms script. If it selects spike times near the beginning or end of the recording, it will produce an indexing error on line 54. I suggest adding:

curSpikeTimes((curSpikeTimes + gwfparams.wfWin(1)) < 1) = [];
curSpikeTimes((curSpikeTimes + gwfparams.wfWin(end)) > size(mmf.Data.x,2)) = [];

after line 49.

Thanks again for making and sharing this code!

~Nick

[nsteinme]

Thanks for catching this. Want to make a pull requested with a tested implementation?

…

On Wed, Dec 6, 2023 at 8:31 AM nmtimme ***@***.***> wrote: Hi, Thanks for making this great resource! I found a bug in the getWaveForms script. If it selects spike times near the beginning or end of the recording, it will produce an indexing error on line 54. I suggest adding: curSpikeTimes((curSpikeTimes + gwfparams.wfWin(1)) < 1) = []; curSpikeTimes((curSpikeTimes + gwfparams.wfWin(end)) > size(mmf.Data.x,2)) = []; after line 49. Thanks again for making and sharing this code! ~Nick — Reply to this email directly, view it on GitHub <#21>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABZ5IP6KK2US74GGECL7FNTYICMWLAVCNFSM6AAAAABAJVCZQ2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGAZDQOJUHAYTGMQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[nmtimme]

Hi Nick,

I've created the pull request (#22 ). I didn't see any tests that could be run on the new code, but if I just missed them, please let me know and I will run them in the fork. Thanks!

~Nick

[Andrea3v]

Just want to report that l also found this issue and corrected it with a similar approach after line 49:

    indx1           = curSpikeTimes+gwfparams.wfWin(end)<nSamp;
    indx2           = curSpikeTimes+gwfparams.wfWin(1)>=1;
    curSpikeTimes   = curSpikeTimes(indx1&indx2);     % skips the spike times whose waveform would fall outside the range 1:nSamp 

thanks
~Andre