Meeting Notes
- Date: 2025-10-07
- Time: 09:00 (PT)
- Location: Teams Meeting
- Presentations: SLAP2 data collection and analysis
Agenda
- Update on data collection plans for SLAP2 data.
- Analysis update 1 - Jessie(Yujie) - 马雨洁 -
- Analysis update 2 - Lucas Rudelt - @lrudelt
Meeting Recording
Meeting Notes
SLAP 2 Platform Data Collection and Experimental Design Updates:
Jerome provided a comprehensive update on the SLAP 2 platform data collection efforts, outlining recent progress, the experimental plan for the next six months, and the rationale for focusing on sensory motor context cohorts, with contributions from Andrew, Madi, and Kaspar.
Recent Data Collection Progress: Jerome explained that the team, including Andrew, Madi, and Kaspar, completed the SLAP version two experiments, resulting in data from approximately ten neurons. This dataset will be packaged and shared with the group to support ongoing analyses, with the possibility of collecting more data if needed.
Experimental Plan for Next Six Months: Jerome described the plan to run two cohorts each for two-photon and neuropixel data, focusing on sensory motor coupling context. The SLAP 2 platform will mirror these designs, enabling cross-comparison between different recording modalities.
Glutamate and Voltage Imaging Strategy: The team intends to conduct glutamate imaging on five mice, imaging the same cells across four sessions (totaling 20 sessions), followed by voltage imaging across dendrites using a recent sensor. Due to platform limitations, glutamate and voltage imaging cannot be performed simultaneously.
Discussion on Calcium Imaging: Lucas raised the possibility of including calcium imaging as a third sensor. Jerome and Kaspar clarified that calcium imaging will be performed in somas with the mesoscope, but technical challenges prevent simultaneous dendritic calcium and glutamate imaging on SLAP 2. Kaspar noted recent improvements in calcium imaging during glutamate imaging but cautioned against promising its inclusion.
Data Availability and Upload Status: Lucas inquired about the availability of August session data, specifically the 25th of August session, and whether missing uploads were due to packaging delays or experimental issues. Jerome confirmed that feedback on data issues is welcome and that further follow-up will occur.
Principal Component Analysis of SLAP 2 Data:
JJ presented a preliminary analysis using principal component analysis (PCA) on SLAP 2 data, detailing the methodology, main findings, and ongoing work, with active discussion and feedback from Jerome, Alex, Lucas, and Nicholas.
PCA Methodology and Rationale: JJ explained that PCA was chosen to reduce the high dimensionality of SLAP 2 data, allowing identification of dominant population-level patterns. The analysis involved flattening the ROI-time matrix and running PCA to extract components, loadings, and scores.
Key Findings from PCA: JJ reported that the first five principal components captured the majority of variance in the data, with PC1 reflecting global activity, PC2 showing a mirror image of PC1, and PC3 and PC5 exhibiting distinct temporal patterns. PC5, in particular, appeared to encode a flip in activity patterns across experimental blocks.
Analysis of Component Trends and Orientation Tuning: JJ described how PC3 and PC5 showed systematic changes across trials, with PC1 and PC2 remaining stable. Further analysis indicated that PC1 captured orientation tuning, with stronger responses to specific stimuli, and PC5 reflected reconfiguration of population responses across blocks.
Feedback and Suggestions for Further Analysis: Jerome, Alex, and Lucas provided feedback, suggesting separation of oddball and standard trials, comparison of PCA components across imaging planes, and visualization of component weights. JJ acknowledged these suggestions and indicated plans to incorporate them in future analyses.
Related Analyses and Collaboration Opportunities: Nicholas mentioned conducting similar PCA-based analyses on orientation tuning and offered to share results for comparison. Jerome and Sarah suggested further exploration of ROI contributions to specific components, particularly PC5.
Contextual Modulation of Synaptic Responses:
New Data Analysis: Lucas and Levke Peter presented new analyses on contextual modulation of synaptic responses using the latest SLAP 2 data, highlighting experimental design improvements, preliminary findings, and next steps, with input from Jerome and the group.
Experimental Design and Rationale: Lucas outlined the updated stimulus paradigm, which includes additional control and oddball blocks to mitigate transient effects and allows the same stimulus to be studied under random, redundant, and deviant contexts.
Aggregate and Temporal Response Patterns: Levke Peter presented aggregate plots showing lower activity for redundant stimuli and heightened activity for deviant stimuli across different orientations and imaging planes. Temporal evolution analyses indicated stable responses in control blocks and specific patterns of depression or enhancement depending on the stimulus and block.
Analysis of ROI Responsiveness: The team analyzed synapses responsive during different control blocks, observing depletion effects for certain orientations and block-specific depression in activity, particularly for the 90-degree stimulus in the third control block.
Data Availability and Technical Issues: Lucas and Levke Peter noted that only one experiment (25th August) was available for analysis due to pending uploads of other sessions. They also reported weak or noisy responses in VIP neuron recordings and discussed possible alignment issues.
Next Steps and Collaboration: Lucas proposed further analysis of individual ROIs and their spatial distribution, as well as collaboration with other groups (e.g., Sarah, Nicholas) to integrate more refined analyses and compare contextual modulation effects across datasets.