Related Experiment Video
Updated: Jun 12, 2025
12:49
A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
Published on: September 28, 2019
12.7K
On human nanoscale synaptome: Morphology modeling and storage estimation
1Sano Centre for Computational Personalised Medicine, Kraków, Poland.
Plos One
|September 25, 2024
Summary
Estimating storage for the human synaptome (neural connections) is crucial for neuroscience. This study models nanoscale synaptic data, revealing petabyte-scale storage needs, even for simplified models.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Big Data
Background:
- Understanding the human connectome requires detailed knowledge of the synaptome, which forms neural microcircuits.
- Synaptic architecture is fundamental to cognitive capacity and neurological disorders.
Purpose of the Study:
- To model the morphology and estimate storage requirements for the human synaptome at the nanoscale.
- To introduce and compare three synaptic models (topologic, point, geometric) for data storage.
Main Methods:
- Defined a synapse based on presynaptic and postsynaptic neuron/terminal pairs.
- Characterized terminals by coordinates, radius, and identifier.
- Calculated storage needs based on synapse count, neuron numbers (30-138 billion), and synapses per neuron (1,000-30,000) across three models and their simplified versions.
Main Results:
- Full synaptic models for the entire brain range from 0.21 to 95.22 petabytes (PB).
- Simplified models reduce storage needs significantly, ranging from 0.14 to 51.75 PB.
- The cortex alone requires storage from 86.80 terabytes (TB) to 8.56 PB.
Conclusions:
- The topologic model is sufficient for connectome topology but exceeds current supercomputer storage capacity.
- Frontier supercomputer can handle the nanoscale synaptome for 86 billion neurons with 1,000-10,000 synapses per neuron.
- This work provides the first large-scale storage estimations for the human nanoscale synaptome.
