Early coactivity gave no clear added gain in a ridge test of 6,524 later calcium changes after stimulation in 75 C. elegans worms

We tested whether early activity correlation helps predict later calcium changes after single-neuron stimulation. We used DANDI 001075 version 0.240930.1859 and two adult C. elegans anatomical maps. The new panel has 6,524 changes from 362 trials in 75 worms. It preserves animal, trial, source and target IDs. The first 40% of stimuli supplied inter-stimulus coactivity. The remaining stimuli supplied outcomes. We used raw frames and fixed coverage rules. Five outer animal folds and four inner animal folds kept each test worm out of fitting and penalty selection. The reference ridge rule used neuron identity, wiring, optical measures and current baseline state. Adding green coactivity increased equal-worm mean squared error from 0.610933 to 0.611040. Error reduction was -0.0175%, with a descriptive 95% fixed-prediction animal range of -0.199% to +0.147%. A later check refit the rules under 199 animal-weight draws. Its range was -0.422% to +0.241%, with folds and penalties fixed. All seven planned sensitivity ranges included zero. An added anatomy ablation also gave no clear separate gain from wiring. Independent readers, direct correlations, a second solver, a planted signal and clean repeats checked the calculations. The package has code, source hashes, a trial dataset, fitted rules, predictions, results, figures and repeat instructions. These results concern one scalar correlation rule and a selected raw-frame panel. They do not test a fitted connectome dynamic model or show that coactivity lacks a biological role. Changes after stimulation include ongoing activity and do not isolate the effect caused by stimulation.

C. elegans · structure and function · connectome · calcium imaging · signal propagation · animal validation · coactivity · reproducible dataset