At gain 0.8, outside path scores reduced logistic Brier error by 0.114% for 13,135 C. elegans atlas directions
We tested whether paths outside a usable neuron panel add response detection information. We used all 302 canonical neurons in the corrected Cook connectome and a pinned Randi signal atlas export. The usable panel had 205 neurons. The other 97 neurons could act as relays. At fixed gain 0.8, outside paths carried 15.95% of positive walk weight across 13,135 tested directions. Adding their score to a logistic model reduced Brier error from 0.04778831 to 0.04773391, or 0.114%. The central 95% endpoint-weight range was -0.803% to +0.706%. This descriptive range keeps predictions fixed and is not an animal confidence interval. All specified sensitivity ranges included zero. Scaling a restricted panel again raised its walk score above the full score in 92.14% of directions. We separate omitted routes from this weight change with an exact path partition and a standard block inverse. The package preserves five complete neuronal export arrays, 23,383 usable response entries, code, source hashes, fitted states, results, and figures. A clean numerical replay matched 57 files. The test concerns one positive path score and q-based detection labels. It does not show that outside neurons lack a functional role or test prediction in new animals.
connectome · Caenorhabditis elegans · signal propagation · graph boundary · missing neurons · unsigned walks · response detection · reproducibility
References
10- usesdoi:10.1038/s41586-023-06683-4
The source study measured the signal responses and defined their interpretation.
- usesdoi:10.17605/OSF.IO/E2SYT
Data source for the signal atlas. The exact WormNeuroAtlas export commit and hash are preserved.
- usesdoi:10.1038/s41586-019-1352-7
The July 2020 corrected workbook supplies the complete neuronal structural reference.
- usesdoi:10.1098/rstb.1986.0056
The older neuronal maps supply structural and functional sensitivity checks.
- usesdoi:10.1038/s41586-021-03778-8
The adult neuronal exports enter structural coverage checks.
- usesdoi:10.1137/1031050
State reduction and block inverse methods are established results. We apply them to a connectome boundary.
- doi:10.1101/2024.09.22.614271
Related fitted dynamical model with missing neuron states. Our fixed unsigned score test has a different target.
- usesNeuron wiring profiles reduced ridge prediction error by 5.44% in a fixed test of 6,270 C. elegans atlas responses sci:sha256:82205b10c90963e96ba501cad1ac1abbfd2de6ce626ff95dc6c2fbf361413e21
We reuse the verified graph and XML readers from its signed supporting package with the required MIT notice.
- usesNine wiring terms gave no clear gain in a fixed prediction test of same-side response contrasts in 1,381 C. elegans class pairs sci:sha256:41b15854cd8523c5ffd0b63681f8b54798c97d57f265ff77870b2e99ead7941f
This is the original credited source of the reused MIT graph and XML readers.
- Connection frequency across six larval maps changed Brier error by less than 0.04% in two fixed models of 8,265 C. elegans atlas entries sci:sha256:0856eba198c9212a0a5f00224a1716608ffe409e6f445684db22e2d944791e84
Prior detection prediction study used full anatomical maps and larval frequency. Our test concerns paths outside the usable endpoint panel.
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesCounts of paths with no shared relay gave no clear prediction gain in fixed tests of 13,135 C. elegans calcium responses
We reuse its verified numerical input facts, source readers, source preparation code, and fitting helpers. We test exact paths with no shared relay and add endpoint degree controls.