On human nanoscale synaptome: Morphology modeling and storage estimation.
Level 5 - mechanism / opinion, no new human data
Theoretical computational modeling and mathematical estimation without new empirical human data (graded by design analogy).
PubMed 39321198 · doi:10.1371/journal.pone.0310156
What was done
The author developed a mathematical framework to model synapse morphology at the nanoscale, defining a synapse by presynaptic and postsynaptic neuron and terminal identifiers, spatial coordinates, and terminal radii. Three distinct representations were evaluated in full and simplified configurations: a topologic model (connectivity identifiers), a point model (topology plus 3D coordinates), and a geometric model (topology, coordinates, and terminal radii). Theoretical digital storage requirements were calculated across 72 scenarios combining four whole-brain neuron counts (30, 86, 100, and 138 billion) with three synaptic density assumptions (1,000, 10,000, and 30,000 synapses per neuron) for both the entire human brain and the cerebral cortex.
What was found
For the entire human brain, data storage demands for full (and simplified) configurations ranged from 0.21 (0.14) PB to 28.98 (18.63) PB for the topologic model, 0.57 (0.32) PB to 78.66 (43.47) PB for the point model, and 0.69 (0.38) PB to 95.22 (51.75) PB for the geometric model. For the cerebral cortex, storage requirements ranged from 86.80 (55.80) TB to 2.60 (1.67) PB for topologic, 235.60 (130.02) TB to 7.07 (3.91) PB for point, and 285.20 (155.00) TB to 8.56 (4.65) PB for geometric models. The author noted that while the topologic model captures connectome topology, the full nanoscale synaptome exceeds the capacity of modern neuroscience supercomputers, with systems like Frontier capable of managing an 86-billion-neuron synaptome only up to 1,000–10,000 synapses per neuron.
Why it matters
This framework provides quantitative bounds on the computational infrastructure and big-data storage capacity required to map and analyze the complete human nanoscale connectome.
Limits
The study is entirely computational and theoretical, relying on broad parametric assumptions from prior literature rather than direct experimental validation or imaging datasets. It does not account for biological heterogeneity across distinct brain regions, dynamic synaptic changes, glial interactions, or the raw imaging data volume needed prior to segmentation.
Cited by
- supports The human brain has approximately 86 billion neurons, each contacting an average of 10,000 neighboring neurons.