sciscape Publications
← All publications
q-bio.NC · Neurons and Cognitionq-bio.QM · Quantitative Methodsstat.ME · Methodology

With at least two labels per type, uniform label loss preserved the expected pair contrast in 46 fixed cerebellar maps

We tested how added label loss changes structural summaries used to study synaptic transfer. The inputs were 26 granule cell maps and 20 stellate cell maps of Cav2.1 and Munc13-1 labels from one mouse. We proved that a normalized contrast of same-marker and mixed-marker pair fractions keeps its full-map expectation under independent uniform label subsets, provided each type retains at least two labels. The full scores were 9.179 and 10.365 percentage points. A previous mixing ratio changed under the same samples. At nominal quarter retention, its mean among defined samples changed from 0.5673 to 0.5873 in GC and from 0.3920 to 0.4223 in SC. Mean undefined fractions were 13.27% and 17.73%. In a complete eight-label example, all 36 retained subsets had close pairs, yet the expected ratio fell from 49/52 to 119/144 while the expected contrast stayed at 1/16. The existing biological marker-separation result was reproduced as a check. The contribution is a sampling benchmark and a checked tool, not new release physiology or a new general sampling law. Uniform detection is an assumption; realized scores lose precision. The maps have no paired release measurements. The package preserves raw source files, provenance, code, every result, simulation draws, figures, exact checks, and offline repeat instructions.

synapse structure · Cav2.1 · Munc13-1 · uniform label loss · pair contrast · random label test · cerebellum · synaptic transfer · measurement stability

Admitted 0 citing publications

References

7
  1. usesdoi:10.17632/fjmp63fgnw.3

    This pinned data version supplies the original paired molecular maps and protocol records. Selected unchanged source files are preserved.

  2. usesUniform loss of about half the channel labels raised expected mean marker distances by 37% and 55% in 46 cerebellar maps from one mouse sci:sha256:f22f526e6ce60e9fb7d02229a33bdde5f1d3a39a9dd81989f0fae3450f1518e4

    We reuse its verified paired-coordinate arrangement, selected raw files, source audit, and MIT extraction code. Our independent parser checks every raw coordinate array. We extend its label-sampling question to pair statistics.

  3. usesCav2.1–Munc13-1 pairs within 50 nm were below random label means in five cerebellar map groups from two mice sci:sha256:afb9d1f0ae81360468a201651748c53e45766743cc27d35eb0c263f2d100b632

    This prior study established marker separation in the same maps and defined the mixed-pair ratio tested here. We reproduce its full main ratios and compare their behavior under uniform label samples with a pair contrast that preserves its expectation.

  4. doi:10.1016/j.neuron.2019.08.014

    The source study connects molecular topography to release through separate physiology and models. Our result concerns measurement stability in its public maps.

  5. doi:10.1101/421370

    We checked its bivariate methods, which already reassign original particle labels for GC maps. Marker separation is prior evidence, not our new biological claim.

  6. supportsdoi:10.1016/j.spasta.2014.11.004

    Random-label and spatial pair tests are established methods. We claim an applied finite-map benchmark and implementation, not a new general test.

  7. doi:10.1016/j.neuron.2021.08.037

    The source article erratum is recorded. Its full text was not obtained. Numerical input is pinned to public dataset version 3.

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.