NeuroPixels Mice

[alexpiet]

Here's a big list of behavior sessions. The other info in this table is still being iterated on/validated, but the experiment sessions are the one for which the stage starts with 'EPHYS'.
"\allen\programs\mindscope\workgroups\np-behavior\vbn_data_release\metadata_220429\behavior_sessions.csv"

[alexpiet]

• It seems like there is a mis-match between session_type and equipment_name. For now I'm just looking at sessions where both seem to match
• BehaviorSession.from_lims(bsid) doesn't seem to work. It looks like its not properly querying lims, which is surprising
• Trying to switch to https://github.com/AllenInstitute/AllenSDK/tree/vbn_2022_dev

from allensdk.brain_observatory.ecephys.behavior_ecephys_session import VBNBehaviorSession

• VBNBehaviorSession doesn't have a from_lims()

[alexpiet]

I opened an SDK issue AllenInstitute/AllenSDK#2406

I did correctly pull some training sessions and fit them. However I should wait for thorough QC testing until I look at the ephys sessions, since they might have different data pre-processing.

[alexpiet]

Discussion on 07/27/2022 with Corbett and Sam

• Familiar sessions should be the same
• Novel sessions have both old and new images. Do we just add an additional strategy for new or old images? Or do we have a strategy for "new image task" and "old image task"? How does timing interact with old/new images?
• SDK now supports NP, so I should be able to work on it now. SDK has cache as well. Data will be publically released on August 12th
• They used stricter criteria for handoff ready behavior. Average minimum reward of 120 rewards/session. I suspect that the most extreme timing mice will be removed, but we will still have a variety of behavior
• Idea: Are visual mice more modulated by holdover images?
• neural activity, is 1st wave of activity the same for visual/timing, but the second wave is where we see differences?