Free positive cell gains fitted 1,342 C. elegans WT–unc-31 mean comparisons within 2.96 sample standard error bands

We tested whether positive source and target gains can align WT and unc-31 response maps in C. elegans. The fixed panel had 1,342 directed pairs. Each pair had at least three finite response values per strain. Each mean could change within its own sample standard error (SEM) band. A complete gain fit required at most 2.956170 SEM. A four-entry constraint ruled out a band of 2.9561693 SEM. The sign-only infimum was 2.954562 SEM. AVER to AWBR set this sign bound and had only three mutant values. At a band of three SEM, this direction required a gain ratio below 0.052. A fivefold ratio limit raised the full bound to 3.102897 SEM. The 21 directions detected in both strains had positive means and no calibration cycles. Their exact fit was therefore automatic. The result separates sign limits, gain limits, and observation-graph limits. It does not identify a mutation effect or measured calibration errors. The bands are sensitivity sets, not animal confidence intervals. The source pools experiments and uses different stimulation durations between strains. We preserve code, derived data, source hashes, complete fits, lower certificates, figures, and repeat instructions.

connectome · signal propagation · Caenorhabditis elegans · UNC-31 · calibration · identifiability · interval constraints