How this list is organized
This journal club list is separate from the lesson pages but follows the technical training sequence. Each item is tagged by the most relevant units so teams can schedule readings alongside module progression. Required readings are the minimum shared core. Optional readings and media broaden context for deeper discussion.
Required papers (3)
1) White et al. (1986) - The structure of the nervous system of C. elegans
Link: https://doi.org/10.1098/rstb.1986.0056
Best fit units: 01, 02, atlas-reference
This is the foundational paper that provided the first near-complete wiring map for a simple organism. It shows what is possible when structural reconstruction is pursued rigorously and systematically. In plain terms, it is the historical proof that mapping wiring diagrams can produce biologically meaningful insights.
Facilitator prompts: What evidence in this paper is descriptive vs mechanistic? Which methodological limits would change interpretation if repeated today?
2) Denk and Horstmann (2004) - Serial block-face scanning electron microscopy
Link: https://doi.org/10.1371/journal.pbio.0020329
Best fit units: 03, 04
This paper introduced a key imaging approach that made high-throughput 3D ultrastructural imaging more practical. It is essential for understanding why modern connectomics became technically feasible at larger scales. In journal club, it helps connect preparation and imaging methods to later computational reconstruction demands.
Facilitator prompts: Which acquisition tradeoffs are explicit versus implicit? What modern QA metrics would you add to this workflow?
3) Januszewski et al. (2018) - High-precision automated reconstruction with flood-filling networks
Link: https://doi.org/10.1038/s41592-018-0049-4
Best fit units: 08, 09
This paper is a cornerstone for modern segmentation at scale. It demonstrates how deep learning can dramatically reduce manual burden while still requiring quality control. For learners, it clarifies how algorithm design and proofreading must be treated as one integrated workflow.
Facilitator prompts: Which failure cases persist despite automation? How should FFN-style performance be reported for scientific, not only engineering, validity?
Optional papers (3)
1) Kasthuri et al. (2015) - Saturated reconstruction of neocortex
Link: https://doi.org/10.1016/j.cell.2015.06.054
Best fit units: 02, 03, 05
This work is a major reference for dense ultrastructural reconstruction in mammalian cortex. It highlights both the biological richness and the technical complexity of large reconstructions. It is especially useful for discussions about scale, annotation burden, and interpretation limits.
Facilitator prompts: What decisions in this study are scale-limited? Which parts would be redesigned for current cloud-scale workflows?
2) Harris and Weinberg (2012) - Ultrastructure of synapses in the mammalian brain
Link: https://doi.org/10.1101/cshperspect.a005587
Best fit units: 05, 06, 08
This review gives a practical structural vocabulary for interpreting synapses and nearby compartments. It helps teams align on what features are biologically meaningful versus ambiguous in EM imagery. In training settings, it supports more consistent proofreading judgments.
Facilitator prompts: Which synaptic cues are robust across datasets? Which cues are most susceptible to staining/contrast variation?
3) Bassett, Zurn, and Gold (2018) - On the nature and use of models in network neuroscience
Link: https://doi.org/10.1038/s41583-018-0038-8
Best fit units: 09, atlas-reference
This paper provides a conceptual framework for how models should be used in neuroscience without overclaiming. It is a strong discussion piece for separating descriptive, predictive, and explanatory modeling goals. It is particularly useful when connecting connectomics analyses to NeuroAI ambitions.
Facilitator prompts: Which model claims in connectomics are currently descriptive only? What evidence upgrades them to explanatory?
Relevant videos/media (up to 3)
1) Sebastian Seung TED Talk - “I Am My Connectome”
Link: https://www.ted.com/talks/sebastian_seung_i_am_my_connectome
Best fit units: 01, 02
This talk is a high-level motivation piece that is still useful for orientation. It communicates why wiring diagrams matter in an accessible way for mixed audiences. It works best as a kickoff discussion, followed by more technical papers.
Facilitator prompts: Which statements are motivational rhetoric vs testable technical claims?
2) MICrONS Explorer (interactive dataset platform)
Link: https://www.microns-explorer.org/
Best fit units: 02, 04, 08, 09
This is a practical media resource rather than a single lecture. It lets learners interact with connectomics data and see how analysis questions are grounded in real structures. For journal club, it works well as a live demonstration companion to paper discussion.
Facilitator prompts: What metadata do you need before interpreting any visualization from this platform?
3) FlyWire platform and tutorials
Link: https://flywire.ai/
Best fit units: 02, 05, 06, 08
This platform provides hands-on exposure to segmentation, proofreading, and morphology interpretation. It is helpful for translating reading discussions into concrete annotation decisions. In mixed-skill groups, it gives immediate visual examples that reduce abstract confusion.
Facilitator prompts: Which proofreading decisions are reproducible across annotators, and which require adjudication policy?
Journal-club technical prep checklist
- Assign one person to lead methods critique and one person to lead limitations critique.
- Require each participant to bring one claim, one supporting metric, and one unresolved uncertainty.
- Track discussion outcomes in a short log: method takeaway, reproducibility concern, follow-up action.
- Label all benchmark and performance numbers with publication year and context.
Suggested cadence
- Week A: required paper + one media demo aligned to current unit.
- Week B: optional paper focused on failure modes or interpretation limits.
- Week C: synthesis session using the atlas reference and dictionary terms.