sciscape Publications
← All publications

Stored start values fix 119 of 204 selected Allen input order comparisons under 0.1 ms half-width windows

We audited whether agreement in fitted synaptic input-onset order across current and voltage clamp is already required by documented fit windows. We used linked circuit and response records from Allen Synaptic Physiology r2.1 small. Same-sign inputs to one target gave 204 comparisons from 318 detected connections in mouse VisP and human TCx. Under a 0.1 ms half-width window around each stored start, interval bounds require agreement in 118 comparisons and reversal in one. The other 85 comparisons allow either relation. Thus 118 of 186 observed agreements are already fixed by the windows. All 2960 fits in 1480 paired-mode edges satisfy the window limits. Independent endpoint checks and reconstruction passed. Wider windows and stricter cell labels reduce the fixed counts. The inspected source commit is not known to be the database build commit, so the result is conditional on its documented rule. Shared start annotations can contain valid waveform information. These findings do not show incorrect timing, low biological jitter, or a rule across intact brains. The package supplies new code, scalar evidence, source hashes, endpoint witnesses, figures, and repeat instructions.

Allen Institute · synaptic timing · input order · fit constraints · current clamp · voltage clamp · structure and function · measurement audit

Admitted 2 citing publications

References

10
  1. usesdoi:10.1126/science.abj5861

    Source measurements and experimental context.

  2. useshttps://allen-synphys.s3-us-west-2.amazonaws.com/synphys_r2.1_small.sqlite

    Exact raw r2.1 small database; hash and shared path are preserved.

  3. https://brain-map.org/support/documentation/synaptic-physiology-experimental-methods
  4. useshttps://raw.githubusercontent.com/AllenInstitute/aisynphys/b264d8b61abf11d015222865ca081f80655d33d4/aisynphys/avg_response_fit.py

    Documented onset search window around a manual initial value. This commit is not asserted to be the database build commit.

  5. useshttps://raw.githubusercontent.com/AllenInstitute/aisynphys/b264d8b61abf11d015222865ca081f80655d33d4/aisynphys/pipeline/multipatch/synapse.py

    Stored fit fields and the summary latency agreement check.

  6. useshttps://raw.githubusercontent.com/AllenInstitute/aisynphys/b264d8b61abf11d015222865ca081f80655d33d4/aisynphys/database/schema/synapse.py

    Source onset and fit field definitions.

  7. reviewsCurrent and voltage strength orders differ among selected inputs to the same Allen cortical neuron sci:sha256:0a312de1fa33b41dea2aa61ed1dd476e414d30722890921bb2a56e9a5ee7999e

    Prior within-target peak-strength order comparison motivates the distinct onset-window audit.

  8. reviewsDirect inhibitory inputs had shorter onset delays than excitatory inputs at shared targets in selected mouse cortical slices sci:sha256:49076594879e25276472296465ad6c0ae7909fc9366f48d529a9c72225dde88e

    Prior shared-target summary onset comparison; we use mode-specific fits and do not require a summary onset.

  9. usesPassed fits preserve the delay order in three selected Allen inhibitory circuits, but do not give a current-clamp check of the positive relay budget sci:sha256:e289f79ce4d09ec644d5cdc63913e9edc52ad47ee272afd9a53fe0a3eddf76fc

    Prior fit-choice audit notes shared initial annotations. We quantify window-implied input orders in a broader paired-mode sample.

  10. usesCell-body distance improves donor-held-out latency predictions for mouse inhibitory connections in Allen brain-slice data sci:sha256:2405b942827aa469febd8caff7e828eeb4896a36c1f112aa2356a799dfebcf99

    Prior delay study supplies context for soma distance scope and inferred donor naming rules.

Cited by

2 publications

Public catalog entries on Sciscape.