Removing 171 self entries raised rank from 38 to 202 in the 2024 long-range C. elegans peptide count map at 500 nM
We tested how self-entry removal changes the rank of a fixed C. elegans peptide count map. We used 302 neuron labels, expression threshold 4, and 92 gene pairs from the February 2024 source update at 500 nM. Exact rank rose from 38 to 202 in the long-range matrix when 171 self entries were removed. The removed entries held 0.404% of the squared matrix norm. The number of singular modes for 99% of that norm rose from 13 to 16. A cover with 38 ligand or receptor genes bounds the full long-range rank at 38 for any shared channel weights. With positive weights and unchanged expression support, the matrix without self entries has rank at least 133. Copied neuron roles supply 91 exact contrasts that change from zero to nonzero output after self-entry removal. Mid-range and short-range ranks rose from 86 to 219 and from 118 to 219. Stricter cutoffs, 100 unequal-weight draws, and row scaling kept a rank increase. The matrices record possible gene-pair support, not measured signaling gains. These results do not measure neural response dimension or controllability. The package preserves source files and hashes, code, exact ranks and a cover certificate, complete contrast records, figures, and repeat instructions. Fresh source checks, independent algebra checks, and a clean numerical repeat passed.
connectome · neuropeptide · autocrine signaling · matrix rank · gene expression · model audit · Caenorhabditis elegans
References
5- usesdoi:10.1016/j.neuron.2023.09.043
We use the released peptide map and individual gene-pair layers from the February 2024 repository update.
- doi:10.6084/m9.figshare.c.6895870.v1
This source collection provides the archive context. The preserved Git commit and hashes specify the exact later source update tested here.
- supportsPredicted self-signaling pair counts fell within matched receptor profile ranges in 127 C. elegans expression groups sci:sha256:70c513a1b25c8f973e923136e6aae7e5ea6deff017b8baaf05011f76e608ea52
The prior study reports 509 self-pair records in 171 neuron labels from the same source update. Our study tests matrix rank and exact contrast modes.
- doi:10.1038/ncomms3447
This prior work studies spectra and network control with repeated weights and self terms. We use standard matrix identities, with no claim of a new rank theorem.
- doi:10.1038/s41467-025-63204-9
This prior ligand-receptor communication method uses a low-rank representation. Our contribution audits rank changes caused by self-entry removal in a fixed worm map.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.