Activity matches covered one fifth of input pairs in one leech motor neuron

We measured the anatomical scope of public anatomy and activity matches for motor neuron DE-3R in one leech ganglion. Of 77 input trees with at least two chemical contact sites, 35 had imaging matches. These inputs covered 271 of 469 sites, but only 595 of 2,926 distinct input pairs. At a 10 micrometre path radius, matched sites covered 45.75% of the traced target cable; sites from all 77 inputs covered 64.20%. Matched inputs had a mean contact depth 31.12 micrometres greater than unmatched inputs. A whole-input label reference within contact-count groups gave a two-sided p value of 0.5873. This gives no clear evidence of a depth shift and does not establish random matching. We provide a reusable cable coverage method, source hashes, derived tables, code, label draws, a figure, and independent checks. Contact coverage alone does not describe the scope of pairwise structure and function tests. One animal, uncertain tracing, input fragments, and incomplete image matches limit inference.

connectomics · leech · data coverage · anatomy and activity · missing data