data and receptive Fields figure for Data release paper (need feedback)

[severine2305]

Hi everyone, I started putting together a figure to show ephys data from P3 for the data release paper. Thanks to Alex who gave me some great plots! This is for one mouse, on 1 specific day. A/ shows rasters and stability of recording for probe C. B/ is the number of SUA+MUA /probe. C/ is population PSTH at preferred RF position for probe C. D/ shows examples of RF for each visual probes.
What would you like to see on this figure that is missing? Any ideas? Comments?

[severine2305]

[severine2305]

C are SUA only

[maierav]

Code to some of the figures shown above:
https://github.com/maierav/claupenscope/blob/main/notebooks/rf_sub830794_probes_ABCD_per_probe.ipynb
and:
https://github.com/maierav/claupenscope/blob/main/notebooks/rf_population_001637.ipynb

Towards a population average:
https://github.com/maierav/claupenscope/blob/main/notebooks/rf_population_allprobes_001637.ipynb

[RaabeM]

Dear all,
I am sharing some example results of the RF analysis for the zebra stimulus, based on Neuropixels data. I am showing the average RF alongside the four most significant single-unit RFs for each probe in one session. The color indicates the correlation between a unit's activity and its best fitting wavelet. Note that the whole screen was used for the zebra stimulus, thus the field of view is a bit wider of what comes out of the Gabor analysis.

[skriabineSop]

Hi everyone ! I think it looks quite nice ! It s quite reassuring that all RF are localized on probe C and D which are targetting visual cortex areas, whereas RFs in ACA seem more scattered. It would be nice to have a schematic showing in 3D where the probes are aiming at, with maybe some WF retinotopies for the cortical area as a sanity control, so it s easier to identify which regions is what. Also, did you specifically use sine or cosine wavelets to established these correlation maps ? in this case, maybe it would be nice to specify which one since it s phase dependant ?

[severine2305]

We are working hard on the probe localization, imaging was delayed. We will have more to share in June.

[RaabeM]

The improved results are due to the fact that I was able to enhance the pipeline's performance, making it possible to test a larger parameter space.
As in the original publication, I use the squared sum of both phases to correlate with a unit's activity. Indeed, this is a coarse approximation. It could be interesting to calculate the correlation dependent on the phase. For a fine characterization of the tuning one probably should do it.

[RaabeM]

Would you mind sharing your code? I will try and merge with the code for the static gratings. This will also allow us to pick the same neurons for direct comparison of RF using either technique.

Yes, ofc! Let me check whether everything is readable and create an env. I will share the link here.

[RaabeM]

Here, the repo for the zebra pipeline => github.com/RaabeM/allen_open_scope_rf
I hope most inconveniences are fixed. Please let me know if you have any problems with it.

[RaabeM]

Dear all,
I will provide an update on our RF analysis tomorrow. If you would like to take a look at the slides in advance, you can do so here: Link to Google Slides. Please note, that the slides are still under construction. Under the link you will find a selection of mass plots, which I hope are understandable. If you want to give feedback, that I can include until tomorrow's meeting, I would appreciate that a lot.
See you tomorrow!