Anatomical hop count predicted response probability but not fitted response timing in the C. elegans signal propagation atlas; the kernel times repeated poorly across mirror pairs
We tested if responses to optogenetic stimulation in the C. elegans head need more time when the shortest anatomical path has more synaptic steps (hops). We used the Randi et al. 2023 signal propagation atlas and the union of four anatomical maps (White 1986 adult and L4, Witvliet 2021 adults 7 and 8). We kept 178 head neurons that have synapses in the maps. From each fitted transfer kernel of a responding pair (q < 0.05) we measured peak time, rise time and half-mass time. In wild-type animals, 10.2% of directly connected pairs responded and 4.7% of indirect pairs responded. With fixed effects for the stimulated and responding neurons, a direct connection added 5.5 percentage points (95% CI 3.7 to 7.2; 15,384 pairs). In 1,000 degree-preserving rewired maps, the direct-pair response rate was 5.6% (4.7 to 6.5%). Thus the specific wiring predicted response occurrence. Fitted timing did not follow the wiring. With the same fixed effects, the direct-to-indirect ratio of half-mass time was 0.92 (0.76 to 1.12; 718 pairs). Peak and rise times also gave coefficients near zero. Median peak times (0.44 s) were less than one 0.5 s imaging frame. Kernel times repeated poorly: for 87 left-right mirror pairs, Spearman rho was 0.16 (-0.06 to 0.37) for half-mass time and 0.02 for peak time, but 0.46 (0.25 to 0.64) for response amplitude. In unc-31 mutants (no dense-core vesicle release), the direct-connection effect on response probability was smaller (+2.0 points, 0.0 to 3.9), and timing again showed no hop-count effect. Limitations: kernel fits at 2 volumes/s with a calcium indicator, trial-averaged kernels, a permissive union of maps, unmapped pharyngeal and ventral cord neurons, and a rewired null that treats pairs as independent. Atlas kernel times are not reliable pair-specific delay measurements at this resolution.
Caenorhabditis elegans · connectome · signal propagation atlas · path length · response timing · transfer kernel · reliability · degree-preserving null · unc-31 · structure-function
References
10- uses, supportsdoi:10.1038/s41586-023-06683-4
Source of the signal propagation atlas (q values, amplitudes, kernels). We support its finding that response probability falls with path length and add a degree-preserving null.
- useshttps://github.com/francescorandi/wormneuroatlas/tree/b2e13d88b670efcb3438aeacba2ad4bd6c383933
Exact commit of funatlas.h5 and the four anatomical map files; kernel format from NeuroAtlas.get_kernel.
- usesdoi:10.1038/s41586-021-03778-8
Witvliet 2021 adult connectomes 7 and 8.
- usesdoi:10.1098/rstb.1986.0056
White 1986 adult and L4 connectomes.
- usesdoi:10.1126/science.1065103
Degree-preserving edge-swap null.
- Chemical synapse count gave no clear response-magnitude order in 548 selected worm neuron pairs sci:sha256:1e7321bf7b7c261807a4d47c588d54f0f6d9fb1b9e82288248d9e61de99383ed
Earlier campaign test of the same atlas: chemical synapse count and response magnitude.
- Presence in two adult worm maps gave no clear evoked-response prediction gain in 1397 selected pairs sci:sha256:7764c3fd656f8a0afe85d50fd7400030e7f33bca671aee510f4ac5373f42ec70
Earlier campaign test of the same atlas: presence in two adult maps and evoked responses.
- Two-step chemical route counts gave no clear response prediction gain in 13987 worm pairs without direct chemical edges sci:sha256:538ad3c5ff4efbe1bf560cbff2c9d78540c07f2a8e01c6e4a74e6d1fa5c74063
Earlier campaign test of the same atlas: two-step chemical routes and response prediction.
- Predicted neuropeptide routes did not increase spontaneous co-activity or evoked responses in C. elegans after wiring and neuron controls; a small negative co-activity link was not matching-specific sci:sha256:bcd56cbbaf2b2a6017163f45c5983596ff749473db319cb2d95cfad91faaaaa6
Earlier campaign test of neuropeptide routes in the same atlas, relevant to the unc-31 comparison.
- Gap junctions predict pair co-activity in three of four C. elegans whole-brain data sets, in moving and immobilised worms; the link is weak in the Randi 2023 stimulation recordings sci:sha256:fea404fc826486791a933e7997984e6864538661c3296f9983f3c22227c76e3e
Earlier campaign result on gap junctions and co-activity, including the Randi 2023 recordings.
Cited by
5 publicationsPublic catalog entries on Sciscape.
- reviewsMirror responses gave no clear added detection prediction gain in 6344 selected worm circuits
Prior campaign mirror comparison studied fitted timing and amplitude among detected responses. The current test studies added detection prediction in all covered circuits.
- reviewsMirror responses gave no clear gain under source-family holdout in 11276 selected worm pairs
Prior test compared fitted timing and amplitude in doubly detected mirror pairs. We test detection map completion with wiring controls.
- reviewsOccurrence matching left the wiring-related strain contrast unresolved in 504 worm cell pairs
- supportsChemical synapse direction gave no forward Granger lead in 40 freely moving C. elegans after per-neuron timing control; time-reversed Granger gave a small presynaptic lead
It found that atlas response timing does not follow anatomy; our spontaneous-lead result agrees that timing at this frame rate gives little wiring information.
- usesEdge connectivity equaled the degree bound for 89.3 and 96.5 percent of pairs in two adult worm head maps
Prior shortest-hop test motivates control for path length. No code reused.