Why Map the Brain
Why wiring diagrams are worth the cost: what a connectome can and cannot support as evidence, and how to frame a question one could answer.
Prerequisites: None. Introductory neuroscience helps but is not required.
Connectomics course
Nine lessons in nanoscale connectomics, from why we map the brain through imaging, reconstruction, proofreading and analysis. Each includes a worked example and an exercise.
Start with lesson 1 and continue in order. Each lesson includes a worked example and an exercise to check your understanding. Use the dictionary and reference atlas whenever you need them.
Learning a specific skill? Try proofreading. Teaching a group? Find lesson plans.
Choose a lesson
Allow about 31 hours for all nine lessons and their exercises. Each card gives its own time estimate and prerequisites. The atlas is a reference to consult as you go.
Brain Data Across Scales
The scales of brain data, and how to choose a representation - volume, mesh, skeleton, or graph - for the question you are actually asking.
Prerequisites: Unit 01
EM Prep and Imaging
How tissue becomes trustworthy images: fixation, staining, sectioning, acquisition, and the QA that decides whether a volume is worth reconstructing.
Prerequisites: Units 01-02
Volume Reconstruction Infrastructure
The pipeline from raw tiles to a queryable reconstruction: alignment, segmentation, agglomeration, storage formats, and provenance.
Prerequisites: Units 01-03; comfort reading code helps but is not required
Neuronal Ultrastructure
Reading neuronal ultrastructure in EM: the organelle catalog, the three criteria for calling a synapse, and calibrated confidence tiers.
Prerequisites: Units 01-03
Axons and Dendrites
Telling axons from dendrites with an evidence chain: the four cue families, the exceptions that break the polarity rule, and why direction errors cost more.
Prerequisites: Unit 05
Glia
Recognizing astrocytes, oligodendrocytes, and microglia - and why a glia-neuron merge is one of the most expensive mistakes in a reconstruction.
Prerequisites: Units 05-06
Segmentation and Proofreading
Segmentation errors, the metrics that measure them and what each is blind to, and proofreading as allocation under a fixed budget.
Prerequisites: Units 03-07
Connectome Analysis and NeuroAI
From reconstruction to claims: graph representations, motif analysis, null-model choice, and the bridge to NeuroAI.
Prerequisites: Units 01, 04, 08; basic probability and Python
Reference — consult, do not complete
Atlas Connectomics Reference
A lookup table of the major datasets, tools, and reference figures - consulted when needed, not read in order.
Prerequisites: None; most useful alongside Units 02-04