Wiring profiles reduced error by 1.65–3.10% with both endpoint groups excluded from training in five fixed C. elegans ridge tests
We tested response prediction when both endpoint name groups had no response values in training. We used fixed ridge models and 6,270 calcium response means from one C. elegans signal atlas export. Anatomy and all observation counts remained known. Across five splits, added wiring profiles reduced mean squared error by 1.65–3.10%. The comparison model included neuron, count, geometry, local wiring, and count profile terms. The primary gain was 1.65%. Its central 95% endpoint weight range was -0.54% to +4.73%. This range is not an animal confidence interval. A later control used the same test sets and outer training row counts. It allowed endpoint groups to occur in training on other pairs. Its gains were 3.60–4.68%. With pharynx endpoints excluded before fitting, the gain was 0.61%. The results show a small transfer gain in this fixed atlas panel. They do not establish prediction in new animals or a signal transmission mechanism. The package preserves source provenance, numerical inputs, split membership, fitted states, test predictions, and repeat instructions.
connectome · C. elegans · signal propagation · endpoint exclusion · ridge regression · structure and function · calcium imaging · prediction validation
References
10- usesWiring profiles reduced error by 4.60% after observation count controls in a fixed test of 6,270 C. elegans calcium responses sci:sha256:04992ad78d9cbd9e6f8d2b7915a6159b53153187fe0109abc980f515385005e5
We reuse its verified numerical inputs, ridge model utilities, count profiles, and fixed penalty grid. We reproduce its pair and source exclusion predictions exactly.
- usesNeuron wiring profiles reduced ridge prediction error by 5.44% in a fixed test of 6,270 C. elegans atlas responses sci:sha256:82205b10c90963e96ba501cad1ac1abbfd2de6ce626ff95dc6c2fbf361413e21
It supplied the anatomical profile method and inherited benchmark utilities.
- usesNeuron terms left 86% of the expected cross product between random halves of 6,270 C. elegans response lists after an asinh transform sci:sha256:e2fcb76e66bbd3b6fc0a95dde6d955b0b33aaac2fec9f7b4953d9e5846c5dbf1
It defined the transformed response list benchmark used by the later prediction studies.
- usesResponse matrix completion reduced error by 2.41% beyond a wiring model in a fixed test of 6,270 C. elegans calcium means sci:sha256:b7d9f820ded7d121c7c54675e017cc5553c2d6853303b6ab4c870e0f0f92671a
Its shared cache supplied numerical files and preserved model utilities. We checked their bytes against the signed source package.
- usesdoi:10.1038/s41586-023-06683-4
It supplied the neural activation and calcium response measurements.
- usesdoi:10.17605/OSF.IO/E2SYT
It is the source signal atlas data record. The tested export commit and file hash are preserved separately.
- usesdoi:10.1038/s41586-019-1352-7
Its hermaphrodite map, corrected in July 2020, supplied the tested anatomical profiles and local wiring terms.
- doi:10.1101/2024.09.22.614271
Version 4 reports a different connectome-constrained dynamical model and validation task.
- usesdoi:10.1098/rstb.1986.0056
Older anatomical maps remain in the inherited numerical input and source rebuild.
- usesdoi:10.1038/s41586-021-03778-8
Older anatomical maps remain in the inherited numerical input and source rebuild.
Cited by
4 publicationsPublic catalog entries on Sciscape.
- usesChemical arrow reversal left profile similarity unchanged with equal input and output weights in five C. elegans maps
We use its verified numerical inputs, MIT model utilities, split rules and saved model states. We reproduce its pooled fits and add a chemical reversal audit and separate role models.
- Five image position terms removed a 0.37% wiring gain in a fixed test of 1,225 C. elegans stimuli
This earlier study tested endpoint transfer of pooled atlas means. The present panel preserves animals and stimulus events.
- usesWith 18 candidates, uniform random choice among names absent from target lists gave 15.7% expected source accuracy in 415 C. elegans events
We reuse its exact canonical coordinate fact arrangement.
- usesWiring profiles reduced ridge test error by 0.64% after exclusion of animals and neuron pairs in 108 C. elegans recordings
We reuse its verified complete graph, position, observation-count, and 6,270-direction numerical fact arrangements. We add original recording and event identities.