an introduction to your connectome

March 26, 2025 8:08 AM

The Brain Mappers (youtube, 30 mins) tells the fascinating story of the growing field of connectomics: The study of how your neurons all fit together to make that big juicy brain of yours, from the map of C. elegans, carefully traced by hand, to using big data to diagram the brains of fruit flies and mice. (found on Kristen M Harris' absolutely amazing SynapseWeb site)
posted by mittens (4 comments total)

the big problem with connectomics is that having the circuit diagram doesn't tell you much about function. c.elegans has something like 300 neurons which are connected in a stereotypic fashion, and it's extraordinarily difficult, if not outright impossible, to simulate its actual behavior. and this is one of the simplest possible organisms to study. trying to understand the function of a mouse connectome, with its exponentially higher number of synapses and more idiosyncratic development is about the equivalent of faster-than-light travel.
posted by logicpunk at 8:38 AM on March 26 [4 favorites]

It makes me happy to be a human being, seeing these smart people giving their all.. A beautiful vid.

Thx for posting.
posted by dancestoblue at 4:55 PM on March 26 [1 favorite]

logicpunk — granting insufficient, do you think useless?
posted by clew at 7:50 AM on March 27

not useless. it's another approach that can help get at function in an indirect fashion, e.g., how does functional/anatomical connectivity in <clinical population> differ from a control population, and how do those differences correlate with differences in behavior or cognition or affect or whatever.

that's a pretty standard way of applying connectivity analyses, and it's tempting to make 1-to-1 links between changes in connectivity strength and function, e.g., if individuals with depression have disordered connectivity between anterior cingulate and insula, then that connection is where major depressive disorder lives, never mind that that specific connection is describing a very coarse parcellation of brain regions, and is only one connection in an entire brain network that is full of analytically intractable feedforward and feedback loops.

so connectomics is important, but i'm not convinced it's more important than other approaches for understanding brain function (which i guess is what the goal is). top-down modeling and theoretical approaches are more likely to produce a comprehensive understanding of brain organization and function than bottom-up cataloguing of the anatomical structure.
posted by logicpunk at 12:18 PM on March 27 [1 favorite]