Seven of eight one-site worm graphs had sensory-to-motor rank bounds within typed degree shuffle ranges

We tested how much indirect wiring adds to a structural sensory-to-motor rank bound in eight public C. elegans brain maps. We kept 161 neurons common to all maps, with 57 sensory inputs and 38 motor outputs. With at least one synaptic site per edge, maximum neuron-disjoint route counts were 29 to 38. Direct sensory-to-motor routes gave 16 to 31. Indirect routes raised the bound in every map. Relative to sampled graphs that preserved each cell's input and output degrees for each broad cell type, the observed bounds differed from mean bounds by 0.00 to 2.29 routes. Seven of eight primary bounds and all eight three-site bounds were within the central 95 percent shuffle ranges. Requiring three sites per edge gave bounds of 6 to 31. Route sets and equal-size neuron cuts certify each exact bound. Random linear transfer matrices also attained the bounds. These results concern generic transfer rank with independent edge weights, separate sensory inputs and motor readouts. They do not measure biological signal dimension, control energy, activity or behavior. Partial head coverage, separate animals and uncertain edge rules limit transfer.

C. elegans · connectome · generic transfer rank · sensory-to-motor routes · development · neuron cuts