search

AllenNeuralDynamics / aind-fip-dff

This capsule takes an input raw NWB file of fiber data and creates processed DFF traces
MIT License
0 stars 0 forks source link

Patch for Iso channel column number error #21

Open hagikent opened 18 hours ago

hagikent commented 18 hours ago

The issue is described here. https://github.com/AllenNeuralDynamics/aind-behavior-blog/issues/735

Briefly, large fraction of Iso channel data.csv files have 9 columns instead of 6 (timestamp, ROI0, 1, 2, 3, Blank). Practically, this is not an issue as what happens is having three additional columns with all 0. As real data should be with values >255, the erroneous columns are easy to be eliminated. Screenshot 2024-11-22 at 3 35 14

This issue has been caused by an bug in the following Bonsai code (will fix this shortly). The node should be Activity.val0 instead of Activity. Screenshot 2024-11-22 at 2 56 56

As this is confusing (why 9 columns? why different from other two channels?), we should take care of this before nwb packaging. Also, some pilot data acquired in 2023Late have only 4columns—Timestamp, ROI0, ROI1, Blank. This should be handled by a patch too.

ZhixiaoSu commented 10 hours ago

Great!! Thank you for finding the culprit. Mid-2024 for sure already has 9 columns. Maybe because some bonsai package upgrade?