Some Analysis Progress

[maierav]

DISCLAIMER: I used Claude Code, which runs into context limits with data of this scope. The below passes the sniff test (after catching many issues), but the code has not yet been audited yet.

We discussed comparison figure for our initial NATURE DATA paper that runs the same analyses across the three techniques. The below is a first shot at that, using a single session only. Another limitation is that we do not yet know which NEUROPIXELS was in which area, so I was unable to ensure that all data are from V1.

We already knew that we have reasonable RFs and orientation tuning using SLAP2. The below shows that we get nice results for the 2P mesoscope and the NEUROPIXELS ecephys as well.

We decided on holding off on showing oddball responses in the initial data paper, but I added a first look out of curiosity.

tl;dr: things look promising so far

Colab Notebook:
https://colab.research.google.com/github/maierav/claupenscope/blob/main/notebooks/openscope_pp_comparison.ipynb

RF MAPPING:

ORIENTATION AND DIRECTION TUNING:

INITIAL PEEK AT ODDBALLS:

[RaabeM]

Hi Alex @maierav ,
Thank you for taking the initiative with the analysis! Before looking into your code, I suggest we perform the same analysis with our pipeline and compare the results. I see, that I can take the according session IDs from your notebook.

Regarding your question: For the pilots, the probe locations are mentioned on the website. In your case, they are:
| Probe | Target |
| Probe A | ACA |
| Probe B | Visa |
| Probe C | Visp |
| Probe D | Vislm |
| Probe E | Mop |
| Probe F | ACA Mos |

You can find them here.