| { |
| "protocol_version": "kc_all5177_structural_atlas_v1_20261030", |
| "created_utc": "2026-10-29T18:16:53.770963+00:00", |
| "timing": "Frozen after coverage/tree/source-coordinate and known-type audit, before new full-population branch/contact metrics, entry connection statistics or displays. Retrospective atlas extension, not confirmatory subtype discovery.", |
| "population": "All5177 frozen KC; all existing1,881,678 unique incident synapse pairs,2,261,016 endpoints,1,252,322branches and5177 skeletons. All former statuses preserved; no new clustering or connectome acquisition.", |
| "entries": "Eight exact frozen stable candidates plus nine existing cross-method consensus entries, with distinct complete IDs and protocol/status. Seventeen atlas views are not17stablemodules. Compare member identity rather than numeric ID suffixes.", |
| "parents": "Primary parent=same-side frozen official type. Secondary within-parent matched hemibrain_type from pinned upstream annotation, only when all primary-hit members share one nonempty label; otherwise composition only. Keep full coarse residual and same-mapped-type residual separately. Audit alreadyobserved CM016147/310rightKCab-s andCM022114/311leftKCab-s alongside original147/799,114/844coarseKCab.", |
| "background": "Existing118102 registered whole-brain somata; exact same v783source annotation coordinates checked using[.004,.004,.04]um transforms. All5177 KC agree. Never present soma point cloud as brain surface or missing somata as absent neurons.", |
| "geometry": "Retain exact original branch IDs/segment fractions. Check source binaryhash, vertexrounding, edgeidentity, treeconnectivity, rawbranchlength and projectedendpointerror for everycell. Displayoriginalcontact coordinates, record mappingqualitycontinuously and prior<=2umflag; no newqualitycutoff.", |
| "features": "Per-cell and perbranch input/output counts, occupiedbranches, pathfromsoma-nearest quantiles, mappingerrors, originalmorphology, knowninputclasses/partnercounts. Perentry fullrole directional synapsecount/uniqueedge matrices and exact KCpossiblepair denominator excludingautapses. NonKC or outside-query denominators remainNA. No physiologicalprobability.", |
| "partial_parent_diagnostics": "CM016/CM022: candidate versus coarse residual and samehemibrain-type residual; pairwise soma distances; within-parent directed edgecounts,density; conductance after symmetrizing directedweighted graph, removeautapses. cut=sumAtoB+sumBtoA;volA=2internalA+cut;phi=cut/min(volA,volB). Report against coarse and mappedparents separately, no newnull/pvalues or boundarythresholds.", |
| "representatives": "For9prior entries preserve existing median-cable representative and selectedbranch. For8stable entries nearest median parent-hit cable length,ties lexicalroot; branch max min(in,out),then total,then branchID. Route family priority ALPN,then exactvisual_projection,thenAPL among availableinputs onthatfixedcell and exactMBONoutput; withinfamily selectmax min(legcounts),then total,then lexicalIDs. Selection uses no distances.", |
| "metricsinterpretation": "Known anatomical subtypes may explain apparent coarse-type subdivision; source hemibrain_type is a mapped annotation in this same FlyWire brain, not new independentconnectome replication. Residualdifferences are descriptive. Stable8/965 unchanged; newstable0.", |
| "validation": "Member/statushashes,allrawdirection/coordinates/branchlineage,all5177rawgeometry+lengthreplay,fullsynapse-endpointconservation,entryrole partition+edgecountconservation,known-type denominators,graphcut fixture,root/branch/route query,portablepayload hashreplay.", |
| "seed": 2026103005, |
| "randomness": "No stochastic scientific analysis; deterministiclexicaldisplay and checks.", |
| "evidence_tier": "DESCRIPTIVE_ONLY_EM_ANATOMY", |
| "model_version": "NONE_NO_NEURAL_SIMULATION" |
| } |
