In one C. elegans model with free local loads, an exact DC and sine fit overestimated a gap weight for two chemical decay times
We tested a method to separate electrical and chemical coupling from complete calibrated voltage responses. We used the 302-cell Cook C. elegans contact support. The stable linear model has reciprocal electrical weights, one positive chemical decay time per direction, and unknown cell loads and capacitances. DC and one complex sine response matrix uniquely recover all model parameters. One sine frequency alone leaves 178 independent mechanism changes. Exact algebra and synthetic tests verify these results. Noise tests show large errors at poorly chosen frequencies. A positive mixture of chemical decay times can increase the fitted electrical weight. In an example with two decay times, a model with one decay time matched both full response matrices within 4.4e-15. However, it gave an electrical weight 4.21 times the true value. After all electrical weights were removed, its selected DC voltage prediction had 43.2% error. An extra frequency detected this model error. All functional parameters are synthetic. We did not fit animal recordings or estimate physiological conductances. The package preserves proof, code, data provenance, models, results, figures and offline repeat instructions.
connectome · electrical synapse · chemical synapse · system identification · chemical decay time · model error · C. elegans · intervention prediction
References
4- usesWith known leaks and reciprocal gaps, full voltage responses leave 26 coupling changes unresolved in a Cook C. elegans model with instantaneous chemical transmission sci:sha256:0c1c93618d6713502f3eae4aefb7443ba21acd429acb804f7548932c222ef542
We reuse its verified CC BY neural support tables and extend its known-filter recovery to unknown edge decay times and a two-decay counterexample.
- usesdoi:10.1038/s41586-019-1352-7
The corrected July 2020 Cook workbook supplies the biological neural contact support through the cited SCI tables. Functional parameters in our study are synthetic.
- supportsdoi:10.1016/j.automatica.2017.12.013
It distinguishes network transfer recovery from identification of internal dynamic modules.
- supportsdoi:10.1002/dneu.22410
It describes parallel electrical and chemical pathways and different mechanisms with similar circuit function.
Cited by
6 publicationsPublic catalog entries on Sciscape.
- Eight cells were the minimum for chemical release blocks in a Cook worm model with known leaks and complete voltage responses
This related study uses chemical timing assumptions to identify mechanisms. Our intervention result uses an instantaneous model.
- usesIn a 302-cell worm model with free loads and positive chemical mixtures, exact DC and two sine matrices allowed a 2.31-fold difference in one electrical weight
We extend its positive-mixture bias result to sharp kinetic-range intervals and extend its two-decay counterexample to exact DC and two sine matrices.
- usesIn a calibrated 302-neuron model with one-pole chemical filters and source-shared times, one exact sine response matrix identified electrical and chemical weights
This prior work solves unknown edge times from DC plus one complex sine matrix and gives the one-sine ambiguity count. Our main addition imposes source-shared times and gives a pure chemical anchor rule for recovery from one complex matrix, plus a known-leak real-rate family.
- Full responses do not fix gap removal in a 302-cell worm model with filtered chemical currents and unknown capacitance and gains
This related study separates chemical filters and electrical coupling when responses are calibrated.
- With unit capacitance, full responses leave 41 relative readout gains free in a 302-cell worm model with reciprocal gaps, unknown leaks and input scales, and instantaneous chemical currents
This prior study shows why assumed chemical kinetics matter for coupling and intervention recovery. We state a separate high-frequency condition for unknown gains.
- supportsWith fixed response error, a slow readout filter shifted the best tested frequency from 0.792 to 0.343 in a 302-cell worm model
It showed noise and filter-model limits for electrical and chemical separation. We test joint conditional information with known chemical timing and an explicit readout filter.