12,270 Publications · 71,396 Citations

We tested whether predicted peptide self-signaling has a global count excess in a published C. elegans map. We used 92 known gene pairs and the February 2024 source update. We grouped 302 neuron labels into 127 distinct source expression profiles. Each random assignment exchanged whole receptor profiles among groups with the same receptor gene count. It kept peptide profiles, receptor frequencies, and receptor co-expression fixed. The observed total was 246 cognate self-pair records. The exact reference mean was 252.03. The central 95 percent range from 20,000 assignments was 231 to 273. The two-sided p value was 0.614; the Holm value for two statistics was 1.00. Eighty units had one or more records, compared with a reference mean of 82.06. Cell type, receptor removal, and anatomical class controls gave similar count comparisons. Without receptor count matching, the expected total was 230.33. The observed count lay within both ranges. This fixed-map test did not detect a global excess under the stated profile exchange assumption. It does not test neural activity, prove equivalence, or measure variation between worms. The package preserves source data and hashes, code, all assignments, results, figures, and reproduction commands. Exact rational checks, exhaustive small cases, and a fresh source rebuild passed. This record corrects a related SCI web link in the first version. The data and numerical results match that version.

We tested whether paired innexin expression terms improve gap junction prediction after contact area and neuron controls. We used 18 gene rows from CeNGEN L4 data. We mapped them to 173 neurons in two C. elegans nerve rings, JSH and N2U. Each test fold excluded its full expression-class pairs from the other animal's training data. Across 9,367 test pairs, the paired terms raised Brier error by 1.75% in JSH and 2.79% in N2U. Mean Brier gain was -0.001765. Its central 95% class-weight range was -0.004911 to 0.000966. These fixed-prediction ranges do not measure variation across animals. Planned changes gave no gain in both maps. The innexin score was within the range of 39 other-gene panels matched for detection count. A secondary test removed a different source fold. It used the same mean training size but allowed the tested class pairs. The paired terms then lowered error by 4.59% and 3.71%. The change between validation scopes remained positive under the class-weight check. This result limits the prediction claim for this fixed model. It does not show that innexins lack a role in wiring. One L4 and one adult animal cannot separate age from individual variation. The package has compact input facts, source hashes, code, fitted models, predictions, figures and repeat instructions.

An earlier study (sci:sha256:e4f996cb5252c2ac6c4115a94528c7ba7efec9fd44e13ce5aaf23789feeb2440) showed that the regional pattern of alpha-synuclein pathology in mice is dominated by a term that does not depend on the injection site, and that this term outranks connectome diffusion models when one site is fitted. Here we ask what that term is made of. We scored each of 213 Allen regions by its mean rank of pathology over 18 injection-site-by-time blocks of published mouse data (caudoputamen, nucleus accumbens and CA1 seeds, 1 to 24 months), leaving out blocks in which the region was itself injected. Across the 139 regions affected at least once, the score followed Snca expression at Spearman rho 0.72, region volume at 0.60 and connectome in-strength at 0.52, while out-strength and distance to the injection sites were null. Every p value came from Moran spectral randomization, which builds surrogate maps with the same spatial autocorrelation; the surrogates reproduced the observed Moran's I and gave null ranges as wide as plus or minus 0.57. Volume and in-strength were correlated at 0.77 and contributed little that was unique (partial rho 0.13 and 0.28), while Snca kept a partial rho of 0.67. With cross-validation on spatially contiguous held-out blocks, volume, in-strength and distance reached R2 0.19; adding Snca raised it to 0.59. However, Snca was not identified against the genome: 716 of 4,213 genes survived Benjamini-Hochberg control against the same spatial null, 14 genes correlated more strongly, and removing the first principal component of expression, which alone correlated at -0.66 with the score, cut Snca from rho 0.72 to 0.38 and from rank 16 to rank 65. An independent one-injection-site tau dataset followed none of these predictors, as expected when the seed-independent term cannot be separated from routing. Group means, an ordinal scale with a large floor, three forebrain seeds and region-averaged expression limit the test.

We tested whether predicted peptide self-signaling has a global count excess in a published C. elegans map. We used 92 known gene pairs and the February 2024 source update. We grouped 302 neuron labels into 127 distinct source expression profiles. Each random assignment exchanged whole receptor profiles among groups with the same receptor gene count. It kept peptide profiles, receptor frequencies, and receptor co-expression fixed. The observed total was 246 cognate self-pair records. The exact reference mean was 252.03. The central 95 percent range from 20,000 assignments was 231 to 273. The two-sided p value was 0.614; the Holm value for two statistics was 1.00. Eighty units had one or more records, compared with a reference mean of 82.06. Cell type, receptor removal, and anatomical class controls gave similar count comparisons. Without receptor count matching, the expected total was 230.33. The observed count lay within both ranges. This fixed-map test did not detect a global excess under the stated profile exchange assumption. It does not test neural activity, prove equivalence, or measure variation between worms. The package preserves source data and hashes, code, all assignments, results, figures, and reproduction commands. Exact rational checks, exhaustive small cases, and a fresh source rebuild passed.