With odor gains fitted, shared glomeruli cut mean 0 to 5 s calcium trace test error by 4.0% against plane controls in three mice
We tested whether a traced common input predicts the calcium response time course after response size and sign can be fitted. We used 371 mitral or tufted cells in 67 glomeruli from three anesthetized mice, with 47 or 48 odors and repeated trials. For each cell and odor, training trials fitted a constant and gain for a mean sister-cell template. Separate trials tested the prediction from 0 to 5 seconds after odor onset. Against templates from other cells in the same imaging plane, mean squared error fell by 5.93%, 5.51%, and 0.44%. The equal-animal mean decrease was 3.96%. A label test with group counts fixed within planes gave conditional p = 0.001 from 999 shuffles. A descriptive glomerulus bootstrap range was 2.14% to 6.38%. Gains were small during the odor period alone. In the third sample, a constant response level gave less error than either waveform template. An exploratory check with the same number of nearest cell bodies kept the main direction. The source study already showed sister-cell timing differences; this work tests transfer to separate trials with fitted gains and plane controls. The result concerns slow calcium traces with target training responses available, not spike timing or prediction of unseen cells from anatomy alone. Gain fits do not match template noise, so cleaner common waveforms can also explain the gain. We preserve compact inputs, source hashes, code, all results, and figures.
olfactory bulb · shared input · glomerulus · structure and function · calcium imaging · temporal response · cross-validation · open data
References
3- usesdoi:10.5281/zenodo.15150149
We use version 2 response traces and anatomical group labels. We preserve compact extracts with source attribution.
- usesdoi:10.1101/2025.04.24.650439
We use its traced parent-glomerulus labels, released cohort, and trial definitions. It already reported shared tuning and distinct sister response time courses.
- supportsdoi:10.1038/nn.2673
This earlier study showed shared rate responses and distinct timing in cells with common glomerular input. It sets prior scope for our calcium waveform test.
Cited by
6 publicationsPublic catalog entries on Sciscape.
- With odor means removed, parent glomerular signals reduced test error by 3.4% in a fixed linear test of 269 traced cells from two mice
This related analysis tested sister response waveform transfer. Our outcome is within-odor trial change.
- usesKeeping full cell responses raised mean odor decoder accuracy by 0.96 percentage points over one fitted component in 67 sister groups from three mice
This earlier compact extraction and sister waveform transfer test gives data provenance and related context.
- usesSelected odor response reversals gave positive mean test scores in 315 traced sister cells from two mice
Earlier compact-response extraction lineage. All means and labels were checked against the primary data again.
- usesIn a fixed ridge test in three mice, traced sister cells cut mean calcium prediction error by 23.0% for excluded odor modules
We use its verified compact response tensors, derived soma coordinates, source receipts, and MIT extraction code. We test transfer to excluded odor modules.
- usesAfter plane mean removal, traced sister cells had excess repeat correlation in two of three mouse olfactory recordings
This verified package supplies factual soma coordinates and their source records for later spatial controls. It also gives the earlier response extraction and transfer analysis.
- usesWith odor gains fitted, an apical length timing rule decreased mean calcium test error by 0.039% in sister cells from three mice
We verified this publication and reuse its exact compact response arrays. We add apical path measurements and a timing rule learned without the scored glomerulus.