Discussion: meetings/2025-10-07

[lrudelt]

Please find the slides to our presentation of contextual modulation in the new SLAP2 recordings here: https://docs.google.com/presentation/d/1USRattCPPdc4nFYd0p28ViQ6JMi1sKqKFiyZh5L9Nxs/edit?usp=sharing

Here is the link to the code: https://colab.research.google.com/drive/14nXYftx5O0Rr4onb9eK6CYwOiMWvVs23?usp=sharing

[jeromelecoq]

ah thank you for those slides.

So quick summary: the NWB file is ok. But the VIP experiment has weaker tuning.
Perhaps @rcpeene could run receptive field analysis on those and share it here ?

[Sarruedi]

Hi @lrudelt thanks a lot for the presentation. Could you please share your code to open the files? Several of my students have trouble loading multiple files and it would be fantastic to have an example script that works in your hands.

[lrudelt]

Right, data seemed to be packaged consistently, it was plug-and-play for Levke to run her code on the 2025-07-02 data sets (thanks guys!!) What you describe sounds consistent with what we saw in terms of responsiveness of ROIs to the oddball stimuli for the VIP experiment. We also saw a similarly weak responsiveness for the pyramidal cell that was recorded on the same day, do you think this is also due to weak tuning, or could this also be caused by weak expression levels?

Based on what we saw, subject 796630 seems to have good expression and nice tuning properties. We analyzed the recording from 2025-08-25 (thanks to @maierav for providing code for the preprocessing!), which is the one that is available so far for this subject. We would love to also have a look at the more recent sessions, especially with Calcium signal at the soma. Do you think those could be made available as well?

Moreover, from the experimental notes it seems that subject 801381 does also have good expression levels, so when these recordings become available we would love to incorporate those as well.

[Sarruedi]

@lrudelt @maierav can you please send the link to the code you mentioned once more? I suspect we are using the wrong notebook at this stage.

[lrudelt]

This is the link to Alex' code for the new stimulus paradigm: https://colab.research.google.com/drive/1GWLAR4cT_wKPheIDFXpe74XezUu6uYL6?usp=sharing

Tomorrow or so we will also provide a link to our code in the post with the slides.

[jeromelecoq]

As you confirm your result, @lrudelt, should we start diving into spatial distribution of those modes?
It seems like there is some interesting overlap with the spatial distribution analysis done with few others.
It would be value to know where this contextual adaptation occurs?

[lrudelt]

Right, the next step is to analyze individual ROIs as well as the spatial distribution of response modulation. We will soon have a plot of the spatial distribution of contextual modulation for individual ROIs, as well as of ROIs that show a change in control responses for the orientations that are shown during the oddball conditions.
We would love to team up though to relate these findings to changes in the entire tuning vectors of individual ROIs, giving us a clearer picture of how contextual modulation relates to stimulus tuning of ROIs, and how this tuning might change in response to oddball stimulation.

[Sarruedi]

I would love to join in on this and am working with lab project students on different aspects. We have just started so it will take a few more weeks, but it could be good to have a chat and discuss how we could approach this from different angles.

[jeromelecoq]

Maybe Jessie can iterate her analysis in collaboration with @Dedalus9 with both DMD as inputs?

[JJYma79]

Yes, I’d be very happy to collaborate with @Dedalus9 and try incorporating both DMDs into the analysis.

Here is the slide: PCA_based Analysis_Jessie.pptx

And scripts：https://colab.research.google.com/drive/1W1AfBoxZAdE5_kxGU9EpOUoEZryjg21i?usp=sharing

[Dedalus9]

Hi @JJYma79! I'll send you an email. Perhaps we can schedule a meeting/call.

[JJYma79]

Thank you very much, Nicholas @Dedalus9 , for sharing your method, and Alex for the highly inspiring discussion.
First, I would like to express my genuine appreciation for your method, your “functional encoding transition mapper” is really impressive.
If you don’t mind, I would like to propose a concrete collaboration step:
I would like to extract the core ROI clusters with the strongest contributions from my PC5 analysis and share this list with you. Hopefully, it may serve as a useful reference

[JJYma79]

Hi! Alex @maierav thank you so much for your guidance again!

Sorry for my delayed reply , your suggestions have been truly valuable. I’ve recently been looking into some papers based on your advice.

Currently, I haven’t yet had the chance to update my previous analysis due to recent workload, but I plan to apply PCA once across both planes very soon and will share the results with everyone as soon as possible.

I also deeply appreciate the hypothesis you raised during the meeting.

Following your suggestion, I looked into several relevant studies. May I ask if these works are in line with what you had in mind? (I just want to make sure I’m on the right track.)：

1.Dhruv NT, Carandini M. Cascaded effects of spatial adaptation in the early visual system. Neuron. 2014 Feb 5;81(3):529-35. doi: 10.1016/j.neuron.2013.11.025. PMID: 24507190; PMCID: PMC3969249.

2.Bastos G, Holmes JT, Ross JM, Rader AM, Gallimore CG, Wargo JA, Peterka DS, Hamm JP. Top-down input modulates visual context processing through an interneuron-specific circuit. Cell Rep. 2023 Sep 26;42(9):113133. doi: 10.1016/j.celrep.2023.113133.

3.Knudstrup SG, Reyes CM, Jensen CM, Schecter RW, Frank MK, Gavornik JP. Oddball Evoked Deviant Responses Reflect Complex Context-Dependent Expectations in Mouse V1. J Neurosci. 2025 Jul 16;45(29):e1859242025. doi: 10.1523/JNEUROSCI.1859-24.2025.

And apologies for a basic question: when you mentioned “shape” in the meeting, did you mean the shape of tuning curves, for example? I’d be grateful for your clarification.

[jeromelecoq]

Here is a link to our current SLAP2 schedule https://docs.google.com/spreadsheets/d/1Y4IDBn6E2i7GtfW9OC9rDWaY_6cWjPxQByVV55THrGE/edit?usp=sharing