Session Kit: Module 10: Network Science and Graph Representation
Everything needed to run Module 10 as a taught session: prep, timing, materials, misconceptions, rubric.
Generated from modules/module10.md. Edit the module page, not this file.
At a glance
| Duration | 4 hours |
| Capability target | Build one connectome graph representation and justify two metric choices for a defined hypothesis. |
| Learners leave with | Graph statistics summary table |
Before you walk in
- You can state the capability target in one sentence without reading it.
- You have one worked example you will narrate, including where you are unsure.
- Data access works — accounts, viewer, notebook — verified today, not last week.
- The rubric is visible to learners before they start, not after.
- You have decided what “uncertain” earns, and you will say so out loud.
Learners should arrive having covered:
- Modules 01-09
Materials
Run of show
| Time | Segment | Your note |
|---|---|---|
| 00:00-08:00 | Graph abstraction choices | |
| 08:00-20:00 | Graph build demo | |
| 20:00-34:00 | Metric computation | |
| 34:00-46:00 | Interpretation and null concerns | |
| 46:00-60:00 | Competency check |
The activity
Scenario: You have the connectivity graph of 500 neurons in a cortical column from the MICrONS dataset. Your PI asks: “Is this circuit small-world? Are there hub neurons? Are there communities?”
- Load the graph and compute basic statistics (nodes, edges, density, components).
- Compute: degree distribution, clustering coefficient, average path length.
- Compare to degree-preserving random graph and Watts-Strogatz small-world reference.
- Identify candidate hub neurons (top 5% by degree or betweenness centrality).
- Run community detection (Louvain or Leiden). Do detected communities align with cell types?
- Write a 1-page graph analysis report with figures, metrics, null comparisons, and biological interpretation.
What learners hand in
- Graph statistics summary table
- Degree distribution plot (log-log scale)
- Community detection results with cell-type comparison
- 1-page report
Misconceptions to target
These are the errors this session exists to prevent. Surface them in the debrief rather than pre-empting them in the lecture — a misconception a learner has voiced is far easier to correct than one they are holding silently.
- They may believe: The connectivity graph is the data, rather than one lossy projection of it that discards all geometry.
- Surface it by asking: “What would have to be true for that to hold? What would change your mind?”
- They may believe: The synapse threshold is a technical detail that does not need reporting.
- Surface it by asking: “What would have to be true for that to hold? What would change your mind?”
- They may believe: A graph metric means the same thing biologically as it does in its original network-science context.
- Surface it by asking: “What would have to be true for that to hold? What would change your mind?”
- They may believe: Erdos-Renyi is an acceptable null for a spatially embedded, degree-heterogeneous connectome.
- Surface it by asking: “What would have to be true for that to hold? What would change your mind?”
Naming the norm
Every session is a chance to make one piece of the hidden curriculum explicit. Pick a moment where you would normally just do the professional thing, and say out loud why you are doing it — then ask whether anyone was taught that.
For this session, the candidate is whichever norm the activity most depends on: stating an assumption in the same sentence as the claim, recording the version a number came from, or saying “uncertain” and having it count as a real answer. See the hidden curriculum for the collected set and why naming them is a fairness intervention rather than etiquette.
Assessment
- Minimum pass: Coherent graph model and metric rationale. Null comparison included.
- Strong performance: Clear link between each metric and a biological question. Multiple null models tested. Community structure validated against external data.
- Common failure to flag: Metric dumping without hypothesis alignment — computing every metric available without explaining what question each answers.
Grade the reasoning, not the answer. A correct call with no evidence chain should not outscore a well-reasoned incorrect one — and saying so publicly changes behaviour within one session.
Exit prompt
State one reason a graph metric might be misleading in your current dataset.
If this session goes wrong
- Nobody talks in the debrief. You asked “any questions?” Ask instead: “Which cue would you drop first if the data got worse?”
- Everyone finishes early. They are pattern-matching, not judging. Give an ambiguous case where the answer is “uncertain” and see what happens.
- Nobody finishes. The scaffolding came off too fast. Work the next case together rather than pressing on.
- A learner is silently lost. The most likely cause is unstated vocabulary. Point them at the dictionary and check back.