• Date: 2026-08-04
• Time: 09:00AM (PT)
• Location: Teams Meeting

Agenda

@jeromelecoq will introduce a migration of our data release manuscript to Myst. The content of the Google doc was migrated and extended to add interactive plots allowing for more in depth understanding of the data. See here the Demo and core GitHub repository

Meeting Recording

.

Meeting Notes

Migration of Data Release Manuscript to MYST: Jerome led a discussion on migrating the data release manuscript from Google Docs to the MYST publication format, explaining the rationale, benefits, and technical details of the transition, with input from Carter, Severine, and Alexander.

Rationale for Migration: Jerome explained that the migration to MYST was motivated by the need for a centralized, standardized, and interactive manuscript format that facilitates collaboration across a globally distributed team and supports downstream analysis and visualization.

Migration Process: The manuscript content was migrated from Google Docs to Markdown using AI assistance, with figures and their source files (Illustrator, PowerPoint, Python scripts) consolidated into a single repository, making it easier to manage and update.

Benefits of MYST Format: Jerome highlighted that MYST, developed by the Jupyter Book Foundations, allows for interactive components, easy integration of code, and static PDF generation, enhancing both the authoring and reading experience.

Interactive Figures and Documentation: The new format supports embedding JavaScript-based interactive figures, collapsible methods sections, and direct links, enabling readers to explore data and methods at varying depths of detail.

Transition Challenges and Collaboration: Alexander raised concerns about the loss of Google Docs' commenting features, and Jerome acknowledged the limitation, noting ongoing discussions with the MYST team and suggesting interim solutions such as using Google Docs for drafts and GitHub pull requests for finalized content.

Data Organization, Visualization, and Access: Jerome, Severine, and Carter described the structure of the data release, including session summaries, interactive tables, and access to raw and processed data, with a focus on transparency and usability for analysis.

Session and Data Summaries: Severine presented a summary figure of all sessions across modalities (neuropixels, mesoscope, SLAP2), detailing session types, chronological order, and quality control, with color coding for different contexts and failure cases.

Interactive Data Tables: Carter explained the inclusion of interactive tables for animals, sessions, and data access, noting that these are synchronized with public spreadsheets and provide metadata, filtering, and direct links to data assets on S3 and DANDI.

Intermediate and Raw Data Access: The data access table includes links to intermediate assets (e.g., spike sorting, segmentation) stored on S3, with explanations of their relevance for advanced analyses and notes on assets not yet publicly released.

Visualization of Raw Data: Jerome described new figures and interactive views for visualizing raw data from each modality, including probe profiles, microscopy movies, and behavior videos, with options for static and interactive exploration.

Behavioral Data Integration: The manuscript includes synchronized behavioral videos, running traces, and stimulus alignment, with code provided for extracting and analyzing these data, supporting further behavioral analysis and re-use.

Authorship Attribution and Contribution Process: Jerome introduced a transparent authorship tracking system integrated with the manuscript, allowing contributors to document their roles and contributions via a portal, with Carter and Dan Bierman assisting in its development.

Authorship Portal Functionality: Contributors can log in with their ORCID, document their specific contributions, and have these roles reflected in an interactive credit table within the manuscript, supporting transparency and recognition.

Contribution Workflow: Instructions are provided for cloning the repository, making contributions (figures, text, code), and submitting pull requests for review, ensuring that all additions are reproducible and properly attributed.

Review and Integration: Jerome clarified that new analyses or figures (e.g., from Nicholas or others) should be submitted as pull requests with accompanying code, and their inclusion as main or supplemental figures will be decided collectively.

Ongoing Analysis, Figure Development, and Collaboration Practices: The group discussed ongoing work on analysis figures (e.g., receptive fields, oddballs, orientation tuning), strategies for collaborative editing, and plans for finalizing the manuscript, with active participation from Alexander, Nicholas, and Karim.

Analysis Figure Development: Nicholas and Alexander described ongoing work on receptive field and oddball analyses, with plans to present results and submit pull requests for manuscript integration; Jerome emphasized the need for both exemplary and summary figures.

Collaboration and Editing Practices: The team debated the balance between using Google Docs for early drafts and GitHub/MYST for finalized content, with suggestions for using PDFs for typo review and leveraging GitHub's pull request system for dialogue and review.

Inclusion of Additional Analyses: Alexander proposed including orientation and direction tuning analyses, and Jerome agreed, recommending that figures show both single-session examples and summary statistics across sessions.

Next Steps and Roundtable Planning: Jerome scheduled a roundtable for the following week to review progress on figures and analyses, aiming to finalize the manuscript once outstanding figures (e.g., unit extraction, zebra, oddball responses) are completed.

Technical and Process Challenges in Migration: Alexander and others raised technical challenges related to the migration, such as the lack of inline commenting in MYST and the technical burden of using GitHub, with Jerome and Carter discussing possible workarounds and future improvements.

Commenting and Collaboration Limitations: Alexander noted the absence of Google Docs-style commenting in MYST, and Jerome acknowledged this, mentioning ongoing communication with the MYST team and the possibility of future enhancements.

Technical Barriers for Contributors: Concerns were raised about the need to clone repositories and use Markdown or GitHub for minor edits, with suggestions to use PDFs for typo review and to tag reviewers in pull requests for collaborative feedback.