What Neuroanatomical Facts can be Derived Using an SSRBC Atlas

Table of Contents

Movies of Lathe Microtome cutting and tape collection in action!

SBPSL Full PowerPoint Presentation (Warning large file! *.ppt file is 29Mbytes)

Movie: Piloting down a virtual neuron's dendritic tree using "Dendritic Explorer" test program (49 Second *.avi file, 22 Mbytes)

Dendritic Explorer test program overview slide

Contacts

What Neuroanatomical Facts can be Derived Using an SSRBC Atlas?

As defined here, a Synapse-Resolution Brain Connectivity Atlas would provide only raw voxel images upon which to base further annotation of regions, pathways, and circuits. The ultimate purpose of this high-resolution data is to guide neuroanatomical specialists in mapping out the complete structural connectivity of the brain. Let us call the hypothetical result of this structural mapping a Complete Neural Connectivity Database (after the National Academy of Sciences, National Neural Circuitry Database Initiative).

What would such a database consist of? It could not consist of a listing of all neurons and their particular synaptic connections. Such a huge listing would be almost impossible to create (see discussion of sparse imaging) and even if it could be created such an unstructured listing would be virtually useless in guiding physiological and theoretical explorations into neuronal functioning. On the contrary, a Complete Neural Connectivity Database must explicitly and concisely represent the statistical regularities of the brain’s connectivity using the well-accepted and perfectly adequate vocabulary of traditional neuroscience. (see Single Brain Physical Slice Library Proposal paper for more in-depth discussion)

Below is a listing of what a Complete Neural Connectivity Database should be able to derive from the raw data in a Synapse-Resolution Brain Connectivity Atlas:

	Regions: A delineation of all brain regions (estimated to be 500-1000).
	Pathways: A listing of all interregional projections (estimated to be 25,000 - 100,000).
	Topographic Maps: A description of the statistics and topographic mapping of all interregional axonal projections.
	Neuron Types: An identification of each region’s neuron types (estimated to be 10 – 100 types per region).
	Neuron Morphology: A delineation of each neuron type’s morphology (dendritic subunits, branching patterns, receptor densities, synapse types (from vesicle and synaptic structure), etc.).
	Local Circuits: A statistical description of each region’s local circuit connectivity (projection neurons, interneurons, input fibers).
	Memory Traces: Information on non-statistical variation in synaptic connectivity within each region (i.e. learning induced intraregional variation). Specifically, examples of neuron-specific synaptic distributions of particular neurons in each region. [1]
	No Physiological (functional) Data: A key job of a neural circuitry database is to guide and constrain physiological experiments and theories. [2]

As noted in the final bullet, a Complete Neural Connectivity Database should not overstep its bounds by hypothesizing function; its job is to provide the anatomical scaffolding upon which to “hang” pharmacological, physiological, and functional results.

[1] As an example, a V1 simple cell having a maximum response tuned to a 45^o edge performs this computation by having developed (i.e. learned) a very particular distribution of synaptic connections (and synaptic weights) relative to the input fiber population coming from LGN. Anatomical descriptions of such synaptic weight distributions are almost nonexistent using today's tracing methodologies, but such neuron-specific synaptic distributions are crucial to any computational theory of a brain region. A Complete Neural Connectivity Database should contain several representative examples of such neuron-specific synaptic distributions for every brain region.

[2] For example, a synapse’s structure is often very stereotypical (for example Gray’s Type II synapse having a symmetrical synaptic densification). The statistical distribution of this class of synapses on the dendritic trees of a particular class of neuron is a purely neuroanatomical fact that belongs in the Complete Neural Connectivity Database. However, the particular neurotransmitter type (or types) associated with this class of synapses cannot be determined from 10nm structural data alone and must be correlated with separate pharmacological experiments. In fact, if such pharmacological experiments were to determine that this morphological classification was incomplete (i.e. that different neurotransmitter types are found in synapses having the same structural morphology) then this would call for an update in the morphological classification rules used in the Complete Neural Connectivity Database (perhaps requiring further specification of synaptic structural features to form the basis of the anatomical classification).

Last Updated: 11/09/2003