Datasets:

cyd0806
/

neuronmimic-data

@@ -82,3 +82,4 @@ wu_20261104_structure_function/manuscript_release/exports/manuscript.pdf filter=
 wu_20261104_structure_function/manuscript_release/figures/ExtendedData1.pdf filter=lfs diff=lfs merge=lfs -text
 wu_20261104_structure_function/manuscript_release/figures/ExtendedData5.pdf filter=lfs diff=lfs merge=lfs -text
 wu_20261104_structure_function/manuscript_release/figures/ExtendedData6.pdf filter=lfs diff=lfs merge=lfs -text

 wu_20261104_structure_function/manuscript_release/figures/ExtendedData1.pdf filter=lfs diff=lfs merge=lfs -text
 wu_20261104_structure_function/manuscript_release/figures/ExtendedData5.pdf filter=lfs diff=lfs merge=lfs -text
 wu_20261104_structure_function/manuscript_release/figures/ExtendedData6.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261105_crossbrain_transfer/manuscript.pdf filter=lfs diff=lfs merge=lfs -text

wu_20261105_crossbrain_transfer/METHODS_RESULTS_zh.md ADDED Viewed

	@@ -0,0 +1,24 @@

+# 吴老师要求推进：跨脑结构预测
+本轮落实“代表性子群可在另一套果蝇连接组中复核”，检验的是已有 **KCab-p解剖父类**，不把原32/15/14标签强行迁移。原成员与旧封存结果保持不变；没有新增湿实验。
+## 方法
+两个独立连接组动物：FlyWire 128个与MaleCNS 129个KCab-p。两半球不是四只动物。分别以一脑为训练、另一脑为测试；仅使用入突触，排除MBON来源入突触。保留方向、伙伴类型和归一化根路径位置，断连分量的路径缺失单列，不删细胞或修补树。MaleCNS伙伴身份使用数据自身提供的FlyWire兼容注释；不自行建立亚型或脑区对应。
+目标为MBON输出接点组成，不是活动或感觉功能。目标词表仅来自训练脑，测试中新标签计OTHER。SVD有效词表、中心化、标准化均由训练脑决定。固定主正则α=10；α=1/100全保留为敏感性，不在测试脑选优。每脑每次在细胞内原生脑区或分支内置换伙伴身份；每种199个抽样、两个迁移方向，共796个空模型拟合。
+## 新结果
+| 迁移方向 | 源脑平均基线 | 几何+伙伴GP | 联合GPJ | 区域/分支条件秩p |
+|---|---:|---:|---:|---:|
+| FlyWire→MaleCNS | 0.049278 | 0.083174 | 0.081510 | 0.355 / 0.865 |
+| MaleCNS→FlyWire | 0.072545 | 0.072738 | 0.072316 | 0.840 / 0.825 |
+指标为平方Hellinger误差，越低越好。第一个方向的联合模型明显不及源脑均值；反向相对均值只改善约0.32%。四项条件秩比较均未支持真实身份—位置关联的独特优势。这里的秩p仅是有限条件置换参考，不是独立动物总体的显著性检验。
+MaleCNS有8,868个预测端点缺少根路径，涉及29/129个细胞，均保留。训练外输出标签平均每细胞占比仅0.0595%/0.1021%；另列输入词表覆盖率。跨脑差异同时包含性别、注释、检测和骨架粒度，不能归因于单一生物机制。
+## 可以怎样写入文章
+当前跨脑结果进一步限制结构联合特征的解释，而不是证明新功能单元。归一化路径表征与旧XYZ表征不同，因此不是对旧3.06%增益的同表征复现；仍是回顾性公开数据基准，不称盲法确认。可以报告“严格保持真实结构与对象的计算框架，在同脑条件空模型、跨脑预测及公开生理三个层面暴露功能推断边界”。尚不能将其升级为Nature正刊发现已成立。

wu_20261105_crossbrain_transfer/README.md ADDED Viewed

	@@ -0,0 +1,9 @@

+# Two-animal known-parent transfer and updated manuscript
+This additive release contains the complete 128+129-cell retrospective cross-animal benchmark, all predictions, 796 conditional-null fits, numerical verification, independent complete rerun checks and the updated 23-page English manuscript (5 main and 7 extended-data figures, 9 verified references including the formal 2026 Cell MaleCNS paper).
+`archives/crossbrain_project.tar.gz` and `archives/manuscript_release.tar.gz` preserve workspace-relative paths. Every member was checked against `archive_members.json`. `release_manifest.json` records uploaded object hashes. `publication_status.json` is the verified remote receipt.
+Required source data are in the previously verified `wu_20261104_structure_function/` release, payload commit `6f24982d0375d957c7c3906e81a2c2678c5c4667`. The extension's `metadata/dependency_dataset.json` maps its exact source paths/hashes to that release. Scientific originals retain their source attribution and terms; no uniform third-party license is inferred.
+The joint representation did not show a consistent independent association advantage. Two connectome animals are not four hemisphere replicates; this does not establish a new functional subgroup, natural multisensory synergy or subtype-resolved physiological transfer. The manuscript is an internal draft, not a submitted or accepted Nature article.

wu_20261105_crossbrain_transfer/archive_members.json ADDED Viewed

	@@ -0,0 +1,546 @@

+[
+  {
+    "path": "archives/crossbrain_project.tar.gz",
+    "bytes": 10732300,
+    "sha256": "95cdd52d942be89a4a7feb1445bfad48813d455a9da05ab08ca305726896f4a1",
+    "members": [
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/README.md",
+        "bytes": 1135,
+        "sha256": "64ffb023974fe031e7d01f85cea542d07e9a45886cea7c76c33366b392dd164b"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/REPRODUCIBILITY.md",
+        "bytes": 2023,
+        "sha256": "9552c0e4bee1f9e3ed02fb046cafb7900989c47728cb4daa9dd15cc659a02895"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/code/build_report.py",
+        "bytes": 10561,
+        "sha256": "72cbb7314e6506756e230cf17ff9c905390b9f6e8f0fa322552e4a512522a980"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/code/package_release.py",
+        "bytes": 7045,
+        "sha256": "7938b2cc4d5adbfef878a06c00a1ec03274b1365f82a0b14681fabf1157102e4"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/code/run_transfer.py",
+        "bytes": 15018,
+        "sha256": "1b44f067edd0b692acc0f6dbe0dc63fbe868b5116f9b4fd8594fa46ff6da8ff5"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/code/test_transfer.py",
+        "bytes": 1633,
+        "sha256": "bc68dbe706a471fa2d91a5fb54fc9418c4f53d01af40d1acd6d36f28d1649f0a"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/code/verify_release.py",
+        "bytes": 892,
+        "sha256": "8ecf137fe040b589082050e4fbaa85387fa42830924371be0413f23efc797023"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/figures/crossbrain_transfer.pdf",
+        "bytes": 34154,
+        "sha256": "b41a92d86a1b6f2d4ba3f34b372bfbbec186c67eaed47a138572d663d7fc7be3"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/figures/crossbrain_transfer.png",
+        "bytes": 263186,
+        "sha256": "47eb9d327e2e669ba8121f3bffee4d68ef2b3fb4ba7757058be725fc33e8056c"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/MaleCNS_Cell2026_europepmc.json",
+        "bytes": 56525,
+        "sha256": "4e5754145c39cec714be4ea39be0d709a25e10ac0702046036b041d1cd885794"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/MaleCNS_biorxiv.json",
+        "bytes": 5195,
+        "sha256": "bff68dd50992b8f72d9ab8d3dcd80752dff096771071a2098f785c3a85474aea"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/MaleCNS_crossref.json",
+        "bytes": 44010,
+        "sha256": "046c07afffb0a124a00a0f43abe1f01d382d211127aee5d87505b503d63597ed"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/MaleCNS_europepmc_search.json",
+        "bytes": 155234,
+        "sha256": "676e35491c00f866f0f49e589126da643b51b5d927125932016f391c862ac0ab"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/README_PUBLICATION.md",
+        "bytes": 509,
+        "sha256": "cf46b4d3af6f9b89ff5fe19448b6e449c27351c4414561209c4d6b0361ff347f"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/SOURCE_ATTRIBUTION.md",
+        "bytes": 1193,
+        "sha256": "ce0931a2a487eac106734b415172a613aadc332f09032a10ca52d061506c4e62"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/dependency_dataset.json",
+        "bytes": 1001,
+        "sha256": "a2f93cb3c061ea973cdf81ba86d5e7e83178647807997e975d4cbe96416004b0"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/environment.json",
+        "bytes": 201,
+        "sha256": "ef7fee4772aa616a5c3e48f01d35ee0f5bb55c183a31833c0b81787b40a4bcda"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/figure_provenance.json",
+        "bytes": 316,
+        "sha256": "f3c34a1e50cac41edb30b56f0f9272eae327c4d0828d522be890ffc3236a8b32"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/preprocessing.json",
+        "bytes": 1216,
+        "sha256": "f68dc291033be8bc97720e052514fccd9b12eddaff08ce800618b34e90d4e73b"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/protocol_freeze.json",
+        "bytes": 135,
+        "sha256": "843a81c75fc559e2aa5f22704c96ce61a41f80e504c1df7fe4c4b4d1bbb519e7"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/run_input_freeze.json",
+        "bytes": 1112,
+        "sha256": "66da4036b069e4d20acdb84bc60c6b25e3518940538684e100ac350f879c441f"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/metadata/scientific_file_manifest.csv",
+        "bytes": 3963,
+        "sha256": "4cdc3be78ed108234f8d7daf86530dc809aa24b7bcb27f1115ff2c128161b2a2"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/protocol.json",
+        "bytes": 2906,
+        "sha256": "ebfa8ab6c3b15aa147bae0497c096b7a27dbf1e82c39cce03385fe1d2d64efe7"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/reproduce.sh",
+        "bytes": 1070,
+        "sha256": "949a3afd02a1f52fb4f68144c63f8a32cb96143b1604dbf13e436fa88de8f780"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/STATUS.json",
+        "bytes": 415,
+        "sha256": "d8409dbf6aebaa8c91ba3ce73341730769f551e4222b6b8146fdde2076cfe1c7"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/cell_scores.csv",
+        "bytes": 164851,
+        "sha256": "58b637d35b1dd56455919d2dca7355a244939cbf89a88fb981c849339a0b9582"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/data_QC.csv",
+        "bytes": 316,
+        "sha256": "3b5ab19d7e245a79cea2d2525c172990de1808dd7a63b766af4d2e0965142661"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/hemisphere_descriptive.csv",
+        "bytes": 2138,
+        "sha256": "acd707d74fbbe98d3d05e103804f8e96307bd9bf2c0adc5f505e5cb83c6ef132"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/model_summary.csv",
+        "bytes": 1523,
+        "sha256": "3c3eec3a6c033ad1bc08fce5b2e2f46d58313d369f220585d142c755b7564c3f"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/null_comparison.csv",
+        "bytes": 1127,
+        "sha256": "79798ec4bbe9e55b0763896625b7d92c62325a60186567e9e68cc8ff1592f8da"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/null_scores.csv",
+        "bytes": 52837,
+        "sha256": "c7ffcbd372668572c30292d9f464cc917a2bcbcc0db2e2aeb088429abe2af1b8"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/predictions.npz",
+        "bytes": 10363719,
+        "sha256": "24b43978fd4663d1179bcc5f49254cce185941fdbd150c3ebe5579f48c1ed657"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/results/transfer_coverage.csv",
+        "bytes": 267,
+        "sha256": "4a22fc64a1b8cee1b09d3cf18cbe4a918464d63969f5158811aa141209e16cb0"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/independent_reproduction.json",
+        "bytes": 473,
+        "sha256": "1c138d16856188f59aeeb5eaaba84994fdcb74956f0b7784d52d11d00e911433"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/independent_rerun.log",
+        "bytes": 5062,
+        "sha256": "50cd41e94bf8d1de64553bfb2fc540c8961cc534167d8dc1e811e84380b8192c"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/null_invariants.csv",
+        "bytes": 308,
+        "sha256": "8e6ecde89411964b113e6feb39308a7b2d57db0af27c8d70e97a4bb96d91ab02"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/run.log",
+        "bytes": 2576,
+        "sha256": "6646138162e8a85ff09b1cf978a6026a64158019de2ec085f560b973544cf184"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/run_attempt01_categorical_input.log",
+        "bytes": 2101,
+        "sha256": "aa8c58bac57cc6df8aee3e74fc542754c6b7e42ff7583131706c7ddd474514a0"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/scientific_validation.json",
+        "bytes": 832,
+        "sha256": "d0ce0dd4c71edca5f0416bf418541e9655f91e28657361a15af96523c532839e"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/validation/visual_review.json",
+        "bytes": 252,
+        "sha256": "6164a0dc9c14fb3e7514c3af444a5d5367ab1d6f889a4f3a1681149af3b8714b"
+      },
+      {
+        "path": "projects/wu_20261105_crossbrain_transfer/\u63a8\u8fdb\u62a5\u544a.md",
+        "bytes": 2482,
+        "sha256": "6d4b794ea82d5845f250abe8c14790ffa47add896ae2aafc1009c3392ea2444a"
+      }
+    ]
+  },
+  {
+    "path": "archives/manuscript_release.tar.gz",
+    "bytes": 3987925,
+    "sha256": "e85cb008e74b99bb63d7c76c4769976c2b07fd9f155ad680b9003f726ec1211f",
+    "members": [
+      {
+        "path": "projects/wu_20261104_manuscript_release/CITATION.cff",
+        "bytes": 485,
+        "sha256": "d0c5ae0e293278d83136f7b3d5b3bef3583ea8a85da089e9557db7f611823e21"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/README.md",
+        "bytes": 2243,
+        "sha256": "96382c2548d9609be9d67d4cb73899af8f4bbe54f235ed3bdf360bf3d15b9cb7"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/REPRODUCIBILITY.md",
+        "bytes": 3567,
+        "sha256": "2370062a55baad7a44032a30c7ab827e7f817532024021b983c5fc01489bc502"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/SUBMISSION_READINESS.md",
+        "bytes": 1995,
+        "sha256": "47143d1a5fc608b43f53b556cf19f36dc884c3aaa5b7543aa2e77362d6935797"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/build_manuscript.py",
+        "bytes": 5215,
+        "sha256": "b6c3debe9e6336d561e28569b999d30091ddf69a9e90ffe70716bc0f74ae2855"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/finalize_data_staging.py",
+        "bytes": 2188,
+        "sha256": "5770a823d022c4cdf25c87be5c685c321a1b657f1d06f6908a88a6ef35481a0d"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/package_data.py",
+        "bytes": 1941,
+        "sha256": "d2d8fc4011eb2e21e72615de519e7fdf0d26d6d2b06fd2ce62772d9fff7b540d"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/prepare_release.py",
+        "bytes": 7186,
+        "sha256": "4fe40d69fc9d43ff01ffa658739d3aae4905b5f39cf5d3a1c345d7cf1428f65f"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/publish_hf.py",
+        "bytes": 4734,
+        "sha256": "1abe16f43ea65957655b556675e792913f2c552a38da5bf95b674eaa458a1496"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/restore_data.py",
+        "bytes": 3614,
+        "sha256": "c7c9b4b0a17139367c4b46626de1b0cd1062909fc334616d8c55bf7fc0301f13"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/revise_manuscript.py",
+        "bytes": 21376,
+        "sha256": "b748dbe2384df04f7b9443c60ff5acdd399c5b4b9f27fb067326485843ea89d8"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/stage_release.py",
+        "bytes": 7502,
+        "sha256": "195056566ad4d35b0ed66b8091816a2dd70dfa195154f8ad75cfbef22cc7ba7a"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/code/verify_data_archives.py",
+        "bytes": 2595,
+        "sha256": "e6a1fd4dfcace3c8031497b45d8321153de74e016cd9757da188e42b0842d871"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/exports/main.log",
+        "bytes": 28294,
+        "sha256": "12f29eb54d20d7bb1fb1c8c8ecca3c5b5ed1095bc95248eaaf98559fe6db39d5"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/exports/main.pdf",
+        "bytes": 986389,
+        "sha256": "47615b8ff60c9ea17214c07f94dec412f7cfd8f0e248f30bce0984cbc1b74320"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/exports/manuscript.pdf",
+        "bytes": 986389,
+        "sha256": "47615b8ff60c9ea17214c07f94dec412f7cfd8f0e248f30bce0984cbc1b74320"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData1.pdf",
+        "bytes": 253771,
+        "sha256": "3e384e198bd6a435a0bd6837d5be386fa7884e8fc43fad932ecf62485f414e2d"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData2.pdf",
+        "bytes": 14606,
+        "sha256": "c66556e9a5f71cd8e50bedfdf60e2e6d32e445f2d358731abf95e006e1e3d8b6"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData3.pdf",
+        "bytes": 18265,
+        "sha256": "b980f10f42ab7826601f2b8b1039709b7c9922cdfe8df4aaff1976fc3624f5db"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData4.pdf",
+        "bytes": 17329,
+        "sha256": "48f1e48058ae50633e0fd1242d59e9ad7e3cff4550e2a909437194d1f4e166f6"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData5.pdf",
+        "bytes": 152489,
+        "sha256": "f67a9289dc97c4b93dd7c264cced740630e1d6465d7f7880f5c1d84aebbf833e"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData6.pdf",
+        "bytes": 177520,
+        "sha256": "7db429180caddabc694bd7c34c7426b792a062d16a2a81ac5c809a1a2061e021"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/ExtendedData7.pdf",
+        "bytes": 34154,
+        "sha256": "b41a92d86a1b6f2d4ba3f34b372bfbbec186c67eaed47a138572d663d7fc7be3"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/Figure1.pdf",
+        "bytes": 17184,
+        "sha256": "c85d93e9ad331b5a067b7dda9beb443986fad48daff6db81d45a6b7ef2c84741"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/Figure2.pdf",
+        "bytes": 27864,
+        "sha256": "74857b2c59a447d6988507961add9571a4e20aad0f8f3b8d2cacd1ab58019679"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/Figure3.pdf",
+        "bytes": 20040,
+        "sha256": "76ebcb98df81f6003fb1a583e78705a8dae3bb112ec3e94280988f51fe6ece40"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/Figure4.pdf",
+        "bytes": 18430,
+        "sha256": "7ee4d70316b8ca751c79131c6935489b310e4b5db01dfde45792672d3561bca9"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/figures/Figure5.pdf",
+        "bytes": 18588,
+        "sha256": "ac9f85ad02c1b247634c25fbccfb3ce4c6fd55a18f63747fea17305d0d5f4d07"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/manuscript/main.tex",
+        "bytes": 36486,
+        "sha256": "85558f2872882311e3f832ec971ac1a6b21168ba4320a5ca6daa15245c3cc404"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/manuscript/manuscript.md",
+        "bytes": 33365,
+        "sha256": "fff8a21f84eb1ccb2cf934d15745a3220af38ff1c8e34c4a8a4b3839e9577659"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/DATASET_CARD.md",
+        "bytes": 3154,
+        "sha256": "6462504d909b6940c40fa8be53a5f0641e43626474b856036d8992a8d43417fd"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/RELEASE_METADATA.json",
+        "bytes": 744,
+        "sha256": "3a89fa1dfca64e52c2177dcca3fa6dc6b5a71b98130f218fea79b2bebe4cbb75"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/THIRD_PARTY_TERMS.md",
+        "bytes": 2315,
+        "sha256": "e58bb4cf9a98c8999358a84f70dfa84d0fa50180222c89c1b2222f419a7b28d5"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/data_archive_manifest.json",
+        "bytes": 3086,
+        "sha256": "b87e1909546c083bca9c2e976fd646630ce14dc5e29aea71913e9089c993a60c"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/figure_provenance.json",
+        "bytes": 4605,
+        "sha256": "8b589529e658eca399beebbc5462de3f77aee15a669ec8ab9fd448b2cdc891f4"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/hf_publication_status.json",
+        "bytes": 12016,
+        "sha256": "ab1f15f27be047a1d6d0fabe0649643595e120cef1a621a6d9e2f1e046734de3"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/publication_status.json",
+        "bytes": 3084,
+        "sha256": "9b05b6cd9051a3c8fb8733f98764cbe127bc4edd189a140fcb54fd734a99af7c"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/release_inventory.csv",
+        "bytes": 2475253,
+        "sha256": "6ed5c653c44e738530402dd511d548def7eb06a54b3737395bfe9565a683014a"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/release_inventory_summary.json",
+        "bytes": 1151,
+        "sha256": "33318f20564389df891ef3e9bdff02665ff1f936b813d61690de5f75070b3fa4"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/shared_code_dependency_closure.json",
+        "bytes": 425,
+        "sha256": "7a0f45c184b884bc47892df0f600f33e2a5ebb7b70b110a46bcfd2175f2d0b3d"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/source_retrieval_ledger.csv",
+        "bytes": 82242,
+        "sha256": "5d7f85abf1a6134982b2616e379567f3e5b95c58718afbcde1cb9b1e14d3b87e"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/metadata/verified_references.json",
+        "bytes": 13665,
+        "sha256": "ca9cb2d3d0e08100db11944b2321d6c64a924b8cacad93d146df3185d8905210"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/requirements-research.txt",
+        "bytes": 475,
+        "sha256": "7137c0457ab4b1802b550c3bc1a14388648ae052d316c4e499bbe8e503d3b2cc"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/account_switch_20261105.json",
+        "bytes": 176,
+        "sha256": "956ff505472d4e9a6dadbf5ff24867065d23860ea5c7c1887c58f288689bfd4d"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/code_staging_validation.json",
+        "bytes": 216,
+        "sha256": "f19fc1d7b02e536cfe723bb939c9e5462fb052a58f4e03ae8b70ad19cd0fff12"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/data_archive_validation.json",
+        "bytes": 302,
+        "sha256": "73ee2a352313451ea460311399823dc28cb202c6e495272a0fe4b84374a0a171"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/data_archive_verification.log",
+        "bytes": 1298,
+        "sha256": "7f797e44ff3acb43405f203511687c9b94749d541056f3b7aa79f98297861185"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/data_packaging.log",
+        "bytes": 2750,
+        "sha256": "075e4eab4a6710de4f5115bdb623d253457871dd85aad9b0d8393516083b922a"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/data_staging_validation.json",
+        "bytes": 254,
+        "sha256": "578d287ee21717525d61fa974bf0d690da9b4394b0c72d87ef19a817af35b12c"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/github_account_switch_20261105.json",
+        "bytes": 88,
+        "sha256": "21f2f5f8105445a13b363f30dfc52b6e319ad7fd742b411e025cb128ec8408a2"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/github_push_status.json",
+        "bytes": 341,
+        "sha256": "a3a863d2b607c6b9466099d8e98fded376326e03b1aeb8defd313c3de5554823"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/hf_upload_20261105.log",
+        "bytes": 76069,
+        "sha256": "6b22d90421dd7e8b00a788818399a28e7d3759954ee2eeb2c804a8d6a1c17d0d"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/latex_pass1.txt",
+        "bytes": 8069,
+        "sha256": "1bf7d0ace60b216e96225cd07bff200fb99b14a88959d75add682dca22633527"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/latex_pass2.txt",
+        "bytes": 8069,
+        "sha256": "1bf7d0ace60b216e96225cd07bff200fb99b14a88959d75add682dca22633527"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/manuscript_page01.png",
+        "bytes": 116635,
+        "sha256": "09fc804fdffc5b12cd2bc61a277a694568aad26f5cec7a450ba2b62d9d9b63ad"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/manuscript_page03.png",
+        "bytes": 164607,
+        "sha256": "9f9b84a441ad3c547b309401a6a6fe8b87d9b914aa5aae66af96f0651547c70f"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/manuscript_page05.png",
+        "bytes": 142859,
+        "sha256": "80ca97040bc6c054cd8cd4651abf9db84d0142101e0f8b928a4e9a27fc2d19ed"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/manuscript_page08.png",
+        "bytes": 122151,
+        "sha256": "7506e77e26f428e6cec307810b85b3ca2d2eb494affe57ed81368017f5e0facf"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/manuscript_validation.json",
+        "bytes": 422,
+        "sha256": "a11cadac48be1334094ec235298e2f3e32a82f74fdedd70ec6054ee3805ebf9c"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/restore_tool_validation.json",
+        "bytes": 387,
+        "sha256": "43b06b404b826c6a3fa7a6b4a613df449b370c8a69785221b7770b84be09b32a"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/source_numeric_audit.json",
+        "bytes": 2258,
+        "sha256": "ddf19a22ddbc7ab4196411445392cefb1eb48645068580f480d0772f0565dfb2"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/updated_page03.png",
+        "bytes": 186625,
+        "sha256": "fd7c538c2ad2b4b419ba2bc8c16d45ce29a68ab2d5312618b92aa36ab20a853f"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/updated_page23.png",
+        "bytes": 45051,
+        "sha256": "5c1f4baa1d050cd785ede17c3b11346f6d9b830f4150a186a12e5367e353c157"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/validation/visual_review.json",
+        "bytes": 211,
+        "sha256": "831e6f5ec00f389fc76c8e0c9feac222539b87f0df6a9784d420c93b26b23d20"
+      },
+      {
+        "path": "projects/wu_20261104_manuscript_release/\u5f53\u524d\u65b9\u6cd5\u4e0e\u7ed3\u8bba.md",
+        "bytes": 5507,
+        "sha256": "825e34ac1df881d10955efefa62e79f4ba990480e5e3347a4439733da5a638fa"
+      }
+    ]
+  }
+]

wu_20261105_crossbrain_transfer/archives/crossbrain_project.tar.gz ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:95cdd52d942be89a4a7feb1445bfad48813d455a9da05ab08ca305726896f4a1
+size 10732300

wu_20261105_crossbrain_transfer/archives/manuscript_release.tar.gz ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e85cb008e74b99bb63d7c76c4769976c2b07fd9f155ad680b9003f726ec1211f
+size 3987925

wu_20261105_crossbrain_transfer/manuscript.md ADDED Viewed

	@@ -0,0 +1,163 @@

+# Testing functional specificity in the mushroom body with synapse-resolved models and public physiology
+**Manuscript type:** Research Article draft for internal review
+**Authors and affiliations:** To be confirmed by the research team.
+**Project contact:** ydchen0806; yindachen@mail.ustc.edu.cn. This contact entry does not assign authorship or corresponding-author status.
+## Abstract
+Connectomes reveal anatomical diversity, but its functional specificity requires tests that separate geometry, input identity and physiological transfer. We combined structure-preserving models of Drosophila mushroom-body Kenyon cells (KCs) with conditional null models and public electrophysiology. Across 5,177 KCs, joint identity–position features improved anatomical output prediction by 3.06%, but regional and branch-conditioned permutations reproduced 98.98% and 99.70% of this gain. A two-brain known-parent transfer test likewise lacked a consistent joint-feature advantage over conditional nulls. A frozen 32-KC recurrent network retained real trees, directed synapse locations and connected output neurons. Its binary response supports were identical across cells at each tested conductance, and an individual input family could trigger propagation without a mixture. In seven public Fly-ID holdouts, a fixed parameter implementation had 35.22 mV mean trace error versus 32.72 mV for a training-waveform baseline. Reconstructing the authors’ quality control reproduced all 24 decisions; errors in the four eligible holdouts were 15.82 and 14.19 mV. Source recordings lacked cell-specific KC subtype labels. These analyses support conditional anatomical and dynamical predictions while limiting claims of new functional subclasses, multisensory synergy and physiological transfer.
+## Introduction
+The mushroom body is a sparse expansion layer [1,2] in which Kenyon cells (KCs) combine inputs from diverse upstream channels before influencing mushroom-body output neurons. Connectome reconstructions now make it possible to preserve the physical substrate of these computations at synapse-level resolution. Connectome-constrained models can predict activity and sensorimotor processing when tested against external measurements [3,4]. The corresponding physiological inference is harder: anatomical input classes are not equivalent to sensory modalities, publicly available KC recordings often lack the subtype identity required for a direct comparison, and model parameters can absorb differences in electrode configuration, quality control and animal-to-animal variability.
+Here we follow a structure-first design: preserve neuron trees, synapse locations, pre- and postsynaptic direction, partner identity and network connectivity; compare anatomically motivated input combinations; and test their explanatory specificity with matched null models and public data. An exploratory right-side split of 32, 15 and 14 KCs was frozen before the current functional comparisons. The 14 unresolved cells retain their KCab-p parent annotation and are not discarded or relabelled to improve a result. ALPN, direct visual projection and indirect visual candidate sources are analysed separately. We ask which claims survive this sequence of structural, dynamical and public physiological tests.
+## Results
+### Identity–position associations do not explain the apparent prediction gain beyond conditioned nulls
+We retained all 5,177 KC objects and used 957,554 incoming endpoints as structural predictors after excluding inputs annotated as MBON. The target was a 35-category anatomical MBON output-contact distribution, containing 256,719 output contacts; ambiguous annotation strings were retained. A five-fold split within known type and hemisphere separated fitting from evaluation, and the vocabulary, dimensionality reduction and scaling were fitted within training folds. The primary score was squared Hellinger error, equally averaged over 18 parent classes with at least 25 cells. Five rare singleton classes remained in fitting and individual prediction records but did not enter that primary macro-average.
+At the fixed primary regularization and spatial scale, the known-anatomy baseline had macro-error 0.05807009; adding separate geometry and partner features (GP) reduced it to 0.05383898, and joint features (GPJ) reduced it to 0.05218953 (Fig. 1). This apparent 3.06% gain over GP was largely reproduced by dimension-matched conditional nulls. Regional identity permutations had mean error 0.05220635 and branch-conditioned permutations 0.05219441, reproducing 98.98% and 99.70% of the gain. The real-versus-null advantages were 0.00001682 and 0.00000488, with permutation-rank p values 0.10 and 0.15 (19 null draws; minimum attainable p = 0.05). Regularization and spatial-scale sensitivities were retained across 340 fits.
+The nulls preserve location, direction, neuropil and per-cell partner counts; branch-conditioned nulls also preserve branch-level partner frequencies. They therefore test an association increment beyond these retained margins and feature capacity. The results do not establish that connectivity or location is biologically unimportant. They show that the proposed unique identity–position explanation is unsupported by this benchmark. Same-brain cell folds also do not provide independent-animal replication.
+### Cross-animal anatomical prediction does not show a consistent joint-feature advantage
+We extended the known-parent audit to a retrospective transfer benchmark between 128 FlyWire and 129 MaleCNS KCab-p cells from two connectome animals [8,9]. Each complete animal was used as the source for fitting, with the other as the destination for evaluation; hemispheres were not counted as independent animals. The new representation used normalized rooted path rather than unregistered XYZ coordinates. It is a separate representation test, not a replication of the original XYZ encoding. All cells and disconnected components were retained (Extended Data Fig. 7).
+At the fixed primary regularization, FlyWire-to-MaleCNS squared Hellinger error was 0.049278 for the source-brain mean, 0.083174 for separate geometry/partner features (GP), and 0.081510 after adding joint features (GPJ). In the reverse direction, these values were 0.072545, 0.072738 and 0.072316. The latter GPJ result improved on the source mean by only 0.32%; the former was substantially worse. These errors concern output-contact distributions within one known parent class and are not directly comparable to the all-parent macro score above.
+We generated 199 region-conditioned and 199 branch-conditioned identity permutations within each animal. Across the two directions, 796 null models were fitted with the same preprocessing rules and feature capacity. Conditional lower-error rank references were 0.355 and 0.865 for FlyWire-to-MaleCNS, and 0.840 and 0.825 for the reverse direction. They did not support a unique association advantage and are not population-level p values for independent animals. The destination contained source-unseen input identities averaging 4.53% and 7.04% of input contacts per cell; source-unseen output labels contributed only 0.0595% and 0.1021%. MaleCNS had 8,868 predictor endpoints without a rooted path, distributed across 29 cells. Such reconstruction and annotation differences, together with sex and detection differences, limit causal attribution of the transfer gap. A separate complete recalculation reproduced every prediction array and score table exactly; this is computational reproducibility, not new biological replication.
+### A structure-preserving network retains the frozen cohort and synaptic geometry
+The primary model contains the frozen 32-cell KC cohort (Fig. 2) and two actual FlyWire v783 MBON14 cells. It retains 1,926 internal KC–KC contacts and 144 KC-to-MBON contacts mapped onto real branches. Boundary contacts comprised 62 ALPN, 196 direct visual projection and 1,098 operational indirect visual candidate contacts. Thirty-five annotation/transmitter-conflicted contacts were excluded before this panel. These labels describe anatomical sources and do not establish the sensory selectivity of every candidate. Only anatomically eligible cells received balanced A/V/L external conductance; missing input classes were not imputed. The remaining cells stayed in the recurrent network. A containing-CV release proxy with a 1-ms delay was used for online recurrence. This is a conditional network model: it does not include APL, DPM, DAN or MBON-to-KC feedback, and its pulses are not sensory stimuli.
+At the previously frozen operating points, binary support sets did not separate the 32 cells. At 0.1 nS they were empty for all cells; at 1 nS all 32 cells responded to all seven non-empty synthetic combinations under the operational detector. In the balanced 1-nS condition, A input alone produced propagating events, so the difference between a mixture and single-channel responses cannot be interpreted as a specific multisensory synergy. These results motivated a public-data transfer test. They also distinguish two mixture comparisons. The primary balanced contrast, ABC minus the mean of A, V and L, averaged 0.014173 mV at 0.1 nS and 11.727164 mV at 1 nS across all 32 KCs (Fig. 3). ABC minus each cell’s strongest single response averaged only 0.008422 and 0.025065 mV. Joint mesh/time refinement changed the primary means to 0.014170 and 11.731084 mV, with maximum per-cell changes 0.000183 and 0.022963 mV. Conductance matching equates external conductance exposure, not injected charge, recurrent event count or downstream arrivals. Thus a large mean-single contrast cannot identify an interaction mechanism or establish natural multisensory synergy.
+### Public animal-level holdout exposes limited zero-refit transfer
+The public author repository accompanying a published KC channel-perturbation study [5] contains 24 raw current-clamp files, ten wild-type files and fourteen NaChBac files. For the fixed model benchmark, the published WT8, WT9 and WT10 parameter rows were treated as a training set, while Fly IDs 1–7 were held out. The evaluation protocol was frozen before the held-out waveforms were loaded. Each trace was baseline-subtracted using its own 100–200-ms pre-step median and evaluated at 1-ms resolution from 10 to 999 ms after the step. The primary comparison was the mean of three current-specific RMSE values per Fly ID.
+The fixed WT9 implementation produced 35.216 mV (Fig. 4) mean per-Fly-ID RMSE across the seven held-out files. The empirical waveform mean computed from the training animals produced 32.723 mV. The paired difference was +2.493 mV (95% Fly-ID bootstrap interval, 0.598–3.784 mV; exact two-sided sign-flip p = 0.046875; six of seven Fly IDs had larger WT9 error). A mean of the three published WT implementations performed worse (39.246 mV), and an RC model fitted only to the training animals’ negative steps performed worse still (43.955 mV). The ensemble waveform range covered only 9.03% of held-out time points, so it should not be interpreted as a physiological prediction interval.
+### Source-quality reconstruction changes the usable sample but not the conclusion
+The source paper [5] excluded recordings with input resistance below 0.8 GΩ. We downloaded its formal Dataset S1 and its analysis code, then reconstructed the authors’ single negative-step exponential fit for all 24 files. The reconstructed input resistance differed from the source table by at most 4.41 × 10−6 relative error, and all 24 inclusion decisions matched (Fig. 5). Because the source table does not contain raw-file identifiers, the condition-specific row mapping was checked using the full input-resistance fingerprint and is recorded as an inferred mapping.
+Four held-out files (Fly IDs 2, 4, 5 and 6) passed the authors’ rule. In this quality-passed subset, WT9 and the training-animal empirical baseline had RMSE values of 15.820 and 14.192 mV, respectively. This was specified as a descriptive sensitivity analysis after the all-file benchmark and was not used to select a preferred model or to add a second significance test. The conclusion therefore remains that fixed parameters have not demonstrated reliable cross-animal transfer.
+### SIZ and electrode assumptions are identifiable sources of variation
+The published morphology description and the released implementation assign different section labels to the SIZ. We therefore evaluated four fixed-parameter variants: code tag12 or paper tag13, with or without the published electrode appendage. On all seven files, the tag13-plus-electrode variant reduced the mean error to 28.685 mV; on the four quality-passed files it increased the error to 21.022 mV, compared with 15.820 mV for the fixed tag12/no-electrode implementation. Choosing the post hoc best variant would therefore reverse the interpretation under a different sample restriction. Four 80-pA variants were refined in mesh and time step; the maximum trace RMSE change was 0.03436 mV and the maximum instantaneous difference was 0.24949 mV. These are finite numerical checks, not a global error bound.
+### Published perturbation data constrain interpretation but do not validate the target subtype
+The source Dataset S1 contains 82 cell records across constitutive, adult two-day and adult four-day NaChBac conditions and controls. After the source quality rule, mean maximum spike rates were 99.86 versus 22.29 spikes s−1 for constitutive control versus NaChBac, 68.60 versus 2.75 spikes s−1 for adult two-day, and 60.45 versus 14.11 spikes s−1 for adult four-day conditions. These are source-data reconstructions of the published perturbation, not new experiments (Extended Data Fig. 4). The workbook does not recover an animal hierarchy, so these cell rows were not treated as independent animals. The public driver labels approximately all KCs; the records do not provide a per-cell γKC or KCab-p label. Consequently, these data constrain general KC excitability interpretations but do not validate target KCab-p physiology.
+## Discussion
+This study establishes a reproducible workflow for connecting synapse-level morphology to conditional network dynamics while exposing the conditions under which the physiological transfer fails. The main positive result is methodological: real trees, contact positions, directions and partner identities can be carried through an online recurrent simulation and evaluated against public source data with a frozen animal-level protocol. The main biological result is deliberately narrower. Current public records do not support a claim that the fixed γKC-derived parameterization predicts KCab-p physiology, a new functional subgroup, or natural multisensory synergy.
+The public benchmark also illustrates why model complexity cannot substitute for independent evidence. A parameter ensemble did not improve the animal-level benchmark, and the apparent benefit of changing SIZ or electrode assumptions depended on whether source-quality-passed files were used. A single-cell RC calibration can reduce error for some files while failing at positive current in others. These patterns are consistent with a mixture of morphology, intrinsic conductance, observation and recording-quality variation. They do not identify which component is causal.
+Several limitations concern the available evidence. The public recordings are retrospective with respect to this project, and the raw files do not provide the subtype and connectome identity needed for a direct KCab-p test. The network omits several feedback pathways and uses an operational release proxy. The frozen 32/15/14 split is retained by design, but its 14 unresolved cells remain unresolved. The results should therefore be read as a computational benchmark and a constraint on the claims, rather than as a completed physiological validation.
+Public-data work can advance without new wet experiments. The two-animal transfer benchmark adds an anatomical generalization test but does not establish correspondence for the exploratory 32/15/14 split. Its directional differences show why fitting and evaluating cells within a single connectome does not establish transfer to a different reconstruction, sex or annotation system. Public APL imaging [6], heterologous receptor data and MBON physiology [7] constrain different model components and should retain separate observation models. A subtype-resolved public confirmation set with animal identity and current calibration would support the strongest next physiological test. A model tuned on the present holdouts would require a different confirmation set.
+## Methods
+### Structural atlas, cohort and cross-brain scope
+The structural inventory contained 5,177 FlyWire KC objects, 2,261,016 directional endpoints and 1,881,678 distinct synapses. All identifiers and annotation ambiguities were retained in the source registry. The right-side 32/15/14 split was frozen as an exploratory cohort partition, not used as an independent functional label. The unresolved 14 cells have an existing KCab-p parent identity; no functional label was inferred for them. Real tree forests were preserved, and soma-path distances were recorded as unavailable where disconnected. Known-parent comparisons in FlyWire and MaleCNS are descriptive comparisons between two connectome animals; hemispheres are not counted as separate animals. Multi-connectome annotations and mushroom-body anatomy provide the biological taxonomy [1,8].
+### Structural prediction and matched conditional nulls
+The predictor hierarchy comprised a known-anatomy baseline, geometry and partner marginals, and a joint identity–position encoding. Training-only vocabulary, SVD and scaling prevented preprocessing leakage across five fixed cell folds within known type and hemisphere. The primary geometry/partner/joint dimensions were 32/32/64; regularization alpha was 10 and the spatial grid 40 μm. Alpha 1 and 100 and grids 20 and 80 μm were reported as sensitivities, not selected for the strongest effect. Targets contained only outgoing MBON-contact distributions; outgoing endpoints, output degree and exploratory cohort membership did not enter predictors. Macro squared Hellinger error averaged 18 parent classes with at least 25 cells. Nineteen regional and 19 branch-conditioned identity permutations preserved designated margins and used identical dimensions, folds and regularization. Permutation-rank p values were (1 + number of null scores at least as favourable as observed)/20. The full 340-fit registry and per-cell outputs are available; this remains a single-brain structural prediction test.
+### Cross-animal known-parent prediction
+The retrospective cross-animal protocol was recorded before the new prediction scores were computed. It used all 128 FlyWire and 129 MaleCNS KCab-p cells. The source dataset supplied fitting observations and the complete destination animal supplied evaluation observations. Incoming endpoints annotated as MBON-class partners were excluded. Partners used supplied FlyWire-compatible type labels, with exact class/superclass fallback strings where unavailable; no new type or region crosswalk was inferred. Geometry comprised five equal-width bins of path divided by each cell's maximum rooted-tree path and one missing-path category. The six fixed geometry bins were combined with a 32-dimensional source-fitted partner SVD; the joint block had 64 source-fitted dimensions. Every source-unobserved column was discarded before SVD, and centering/scaling used only source rows. Continuous covariates were log input count, log maximum path, missing-path fraction and unknown-annotation fraction. Targets used source-observed exact MBON annotation strings and an OTHER category for destination-only labels. They were normalized contact proportions, not activity measurements.
+Ridge regularization was fixed at alpha 10; alpha 1 and 100 were retained without destination-based selection. Primary performance was mean per-cell squared Hellinger error over the complete destination cohort. A source-brain mean of output probabilities was the principal competing baseline. Regional and branch-conditioned permutations used native source region annotations within each cell, adding native branch identity for the latter. No region correspondence across brains was needed for these within-brain shuffles. Each draw independently permuted source and destination identity assignments, preserving geometry, conditional group membership and partner marginals, followed by source-only fitting. There were 199 draws per scheme, yielding minimum rank reference 1/200. These are finite conditional references and not independent-animal population tests. The sources had previously been inspected; the protocol is not a blinded prospective preregistration.
+### Connectome and network model
+The frozen cohort, real morphology, synapse endpoints and partner identities were inherited from the sealed structural atlas and combination-network projects. ALPN, direct VPN and operational LVIN candidate contacts were kept as separate labels. Real contacts were mapped to cable branches and merged at the containing computational volume where needed. External conductance balancing used (B = \min(n_A,n_V,n_L)) and scaled only downwards; no missing contacts were synthesized. The channel parameter transfer and candidate channel-localization rule were hypotheses; no conductance was increased to achieve a desired waveform. The recurrent release proxy detected a containing-CV crossing of 0 mV and delivered the event after 1 ms. The main network used KC maximum extent 0.25 μm, MBON maximum extent 0.5 μm and time step 0.0025 ms (158,568 CVs); joint refinement used 0.125/0.25 μm and 0.00125 ms (299,873 CVs). Synaptic conductance regimes were 0.1 and 1 nS per reference contact before downscaling. The combination panel comprised 24 main and 10 refinement runs; computational conditions are not independent biological replicates. The primary voltage window was 100 ≤ t < 1,100 ms relative to the matched no-input trace. MBONs used transferred passive parameters and were not treated as independent physiology.
+### Public electrophysiology benchmark
+Raw ABF files were read with `pyabf`; file paths, SHA256 digests and Git blob identifiers were recorded. The training set consisted of the published WT8, WT9 and WT10 parameter rows. Fly IDs 1–7 were held out for the primary benchmark. The current steps were 40, 80 and 120 nominal pA. Voltage was baseline-subtracted using the 100–200-ms pre-step median and interpolated to 1-ms samples at 10–999 ms after onset. For each trace, RMSE was the square root of the mean squared voltage error over time; these trace RMSEs were then averaged equally over the three currents and finally over Fly IDs. This differs from pooling all squared errors before taking a square root. Ten thousand paired bootstrap resamples (seed 1104) were performed over Fly IDs only; an exact 128-pattern sign-flip calculation was used for the two-sided paired primary contrast. Its inferential interpretation requires symmetric/exchangeable sign errors and is limited to the seven available public files. Public prior access means this local protocol freeze is not a prospective blinded preregistration. The 120-pA condition was retained as a high-drive stress test; a 40/80-pA descriptive sensitivity is provided because the paper’s main spike-waveform analysis used 10–100 pA.
+### Source quality control
+The published MATLAB method was followed for the negative current step: an exponential response was fitted to the first hyperpolarizing current trace, input resistance was derived from the fitted amplitude divided by the nominal current, and recordings below 0.8 GΩ were flagged. The formal Dataset S1 was obtained from Europe PMC; its MD5 matched the checksum in the article XML. The workbook’s lack of direct ABF identifiers is reported explicitly. The source author fit uses the first 200-ms mean baseline and the first negative-step exponential fit, whereas the primary prediction benchmark uses the separately frozen 100–200-ms median baseline. These operations are not conflated. Multi-prominence peak descriptors are retained separately from author-verified spike counts. The ABF current-monitor step is approximately half the nominal DAC/source-paper current; lacking a hardware gain specification, all primary analyses retain nominal currents and do not assert a calibrated physical rescaling.
+### Native implementation diagnostics
+The released author morphology and native NEURON mechanisms were used with fixed WT9 parameters. We tested section labels 12 and 13 and the presence or absence of the published 10-µm electrode appendage. The diagnostics were not used to refit the target model or to replace the primary benchmark. A separate 80-pA refinement changed neuronal SWC mesh target from approximately 1 to 0.5 µm and the integration step from 0.01 to 0.005 ms; the electrode remained one segment.
+### Reporting and statistical scope
+The physiological benchmark used Fly IDs as its inferential units. Structural conditional references have a separate scope; the cross-animal transfer includes only two connectome animals, not four hemisphere replicates. Cell rows in Dataset S1 were used for descriptive source reconstruction only. No claim of independent laboratory replication is made. Natural sensory stimuli, behaviour, APL/DPM/DAN feedback and target KCab-p membrane recordings were not inferred from the conditional network.
+## Data availability
+All nine prior task rounds, the structural atlas and auditable dependencies are published at https://huggingface.co/datasets/cyd0806/neuronmimic-data under `wu_20261104_structure_function/`, scientific payload commit `6f24982d0375d957c7c3906e81a2c2678c5c4667`. All 12 archive objects and 8,850 included members were locally checked; the 71 indexed remote objects were checked by size and LFS SHA256 or Git blob identity. The cross-animal extension, all predictions, source-linked figure and updated manuscript are provided under `wu_20261105_crossbrain_transfer/`; its immutable commit is recorded in the publication receipt. Original-source retrieval requirements for 144 files remain explicit. No blanket license is applied to third-party materials.
+## Code availability
+Analysis code and frozen protocols are available at https://github.com/ydchen0806/bio_fly_cyber_lateralization on `research/wu-structure-function-20261104`; the initial verified release commit is `f6042fe34ec1f9fe454769aa49774171ded0c144`. The cross-animal extension preserves workspace-relative paths and has a separate read-only verifier and isolated complete-rerun entry point. The publication receipt records subsequent verified code and dataset commits. No credentials are included.
+## Reporting summary
+The principal experimental unit for the public benchmark was the Fly ID. There were seven primary held-out Fly IDs and four source-quality-passed Fly IDs in the sensitivity analysis. No animals were newly collected. The public source records are not subtype-resolved at the level required for a KCab-p claim.
+## References
+1. Feng Li et al. The connectome of the adult Drosophila mushroom body provides insights into function. *eLife* (2020). https://doi.org/10.7554/eLife.62576
+2. Ishani Ganguly et al. Diversity of visual inputs to Kenyon cells of the Drosophila mushroom body. *Nature Communications* (2024). https://doi.org/10.1038/s41467-024-49616-z
+3. Philip K Shiu et al. A Drosophila computational brain model reveals sensorimotor processing. *Nature* (2024). https://doi.org/10.1038/s41586-024-07763-9
+4. Janne K. Lappalainen et al. Connectome-constrained networks predict neural activity across the fly visual system. *Nature* (2024). https://doi.org/10.1038/s41586-024-07939-3
+5. Katie Greenin‐Whitehead et al. Ectopic sodium channel expression decreases excitability of Drosophila Kenyon cells. *The Journal of Physiology* (2025). https://doi.org/10.1113/JP288790
+6. Hoger Amin et al. Localized inhibition in the Drosophila mushroom body. *eLife* (2020). https://doi.org/10.7554/eLife.56954
+7. Omar A Hafez et al. The cellular architecture of memory modules in Drosophila supports stochastic input integration. *eLife* (2023). https://doi.org/10.7554/eLife.77578
+8. Philipp Schlegel et al. Whole-brain annotation and multi-connectome cell typing of Drosophila. *Nature* (2024). https://doi.org/10.1038/s41586-024-07686-5
+9. Stuart Berg et al. Sexual dimorphism in the complete Drosophila male central nervous system connectome. *Cell* (2026). https://doi.org/10.1016/j.cell.2026.08.015
+## Figure legends
+**Figure 1 | Testing whether identity–position associations add structural explanatory power.** Fixed known-anatomy, geometry/partner (GP) and joint (GPJ) representations are compared with dimension-matched regional and branch-conditioned identity permutations across 5,177 KCs. Outcome: macro squared Hellinger error for anatomical MBON output distributions, not activity or independent-animal validation. Main null distribution uses 19 draws per scheme. Source: association project `figures/association_controls.pdf` and its linked synthesis tables.
+**Figure 2 | Anatomy-preserving input combinations in the frozen KC cohort.** Real KC trees, directed contacts and the anatomical eligibility for ALPN/directVPN/operationalLVIN input balancing. Ten of the original 32 KCs have all three source classes; other cells remain in the recurrent network without fabricated inputs. The two MBON14 cells receive only the modelled cohort’s true contacts. Source: combination project `figures/anatomical_coverage.pdf`.
+**Figure 3 | Complete primary combination contrast and joint numerical refinement.** ABC minus mean(A,V,L) is reported for all 32 cells and both MBONs at 0.1 and 1 nS, using main and jointly refined meshes/time steps. This is not ABC minus strongest single, a charge-matched intervention or a sensory-synergy assay. Numerical-rest MBON residuals at low weight retain their physical scale. Source: combination project `figures/full_primary_refinement.pdf`.
+**Figure 4 | Zero-refit public Fly-ID holdout with competing baselines.** Seven held-out Fly IDs, each represented by the mean of three nominal-current trace RMSEs. Training empirical mean uses only Fly IDs 8–10. Lines join the same Fly ID; circles indicate inferred source-QC-passed files and crosses source-excluded files. The seven files constitute a within-study retrospective holdout; QC annotations were reconstructed after the primary benchmark. Source: public-validation `figures/animal_model_comparison.pdf`.
+**Figure 5 | Reconstructing source quality control changes the usable sample.** Left, source versus reconstructed input resistance for all 24 ABFs; the dotted line marks 0.8 GΩ. Right, mean trace error in the four source-QC-passed holdouts. This is descriptive sensitivity and is not used for a new p-value or preferred-variant selection. Condition-within-row mapping is inferred and checked by resistance fingerprints. Source: public-validation `figures/author_QC_and_sensitivity.pdf`.
+**Extended Data Figure 1 | All 21 held-out waveforms.** Observed KC voltage, fixed WT9 and training-waveform mean; shared physical voltage axes, all files and currents retained. Source: `all21_heldout_traces.pdf`.
+**Extended Data Figure 2 | SIZ/electrode sensitivity.** Fixed-parameter native NEURON variants, all seven files and four QC-passed files shown separately. No variant replaces the frozen primary model. Source: `native_variants.pdf`.
+**Extended Data Figure 3 | Observation and same-animal calibration diagnostics.** Current-monitor/DAC ratios remain an uncalibrated observation issue; same-animal negative-step RC calibration is a protocol-transfer secondary analysis. Source: `observation_and_calibration.pdf`.
+**Extended Data Figure 4 | Published perturbation source reconstruction.** Eighty-two source cell records, with original input-resistance exclusions retained, displayed by NaChBac expression regime and control. Horizontal lines show retained-cell means; cell-to-animal hierarchy is not supplied. This is published source-data reconstruction, not a new independent experiment. Source: `published_perturbation_source.pdf`.
+**Extended Data Figures 5 and 6 | Raw WT and NaChBac quality panels.** All 24 files retained, nominal negative/positive currents plotted without label-driven exclusions. Source: `raw_WT_QC.pdf` and `raw_NaChBac_QC.pdf`.
+**Extended Data Figure 7 | Two-animal transfer within the known KCab-p parent.** Source-animal fits and complete destination-animal evaluation, including all 257 cells. Top: fixed-alpha model comparisons and missing-path coverage. Bottom: 199-draw regional and branch-conditioned references for each direction and all fixed-alpha sensitivities relative to the source-brain mean. Two animals are not four independent hemisphere replicates. Source-only vocabulary/SVD/scaling and source-unseen labels are reported. This normalized-path representation differs from the original XYZ encoding and does not establish new subgroup homology. Source: crossbrain-transfer `figures/crossbrain_transfer.pdf` and its linked tables.

wu_20261105_crossbrain_transfer/manuscript.pdf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:47615b8ff60c9ea17214c07f94dec412f7cfd8f0e248f30bce0984cbc1b74320
+size 986389

wu_20261105_crossbrain_transfer/release_manifest.json ADDED Viewed

	@@ -0,0 +1,44 @@

+[
+  {
+    "path": "METHODS_RESULTS_zh.md",
+    "bytes": 2482,
+    "sha256": "6d4b794ea82d5845f250abe8c14790ffa47add896ae2aafc1009c3392ea2444a",
+    "git_blob_sha1": "2d737bb39e5c01eb632105904c4390c8d592e483"
+  },
+  {
+    "path": "README.md",
+    "bytes": 1407,
+    "sha256": "86750e7905a53ee0bcd0caf2bfc03e9f9061e6d6dc8d6c7b16236c30d27c5ac9",
+    "git_blob_sha1": "7147653f588b3afc5487a83fe459ea7972efcb63"
+  },
+  {
+    "path": "archive_members.json",
+    "bytes": 22986,
+    "sha256": "3596ce12cc40a5f2c31371f44a0637faaedd6309c32694029daf0a1e6a5b884b",
+    "git_blob_sha1": "4d1c9fda36294da25bfd2a3388bd6f0c792f978c"
+  },
+  {
+    "path": "archives/crossbrain_project.tar.gz",
+    "bytes": 10732300,
+    "sha256": "95cdd52d942be89a4a7feb1445bfad48813d455a9da05ab08ca305726896f4a1",
+    "git_blob_sha1": "5f61f662229e5a3442a93720c4dce2af994e189a"
+  },
+  {
+    "path": "archives/manuscript_release.tar.gz",
+    "bytes": 3987925,
+    "sha256": "e85cb008e74b99bb63d7c76c4769976c2b07fd9f155ad680b9003f726ec1211f",
+    "git_blob_sha1": "a39350b262eb12ec3cca2bf2f677141dd5b94015"
+  },
+  {
+    "path": "manuscript.md",
+    "bytes": 33365,
+    "sha256": "fff8a21f84eb1ccb2cf934d15745a3220af38ff1c8e34c4a8a4b3839e9577659",
+    "git_blob_sha1": "8c8c17d31e4a26f5f195d171d3445c9d8af443cf"
+  },
+  {
+    "path": "manuscript.pdf",
+    "bytes": 986389,
+    "sha256": "47615b8ff60c9ea17214c07f94dec412f7cfd8f0e248f30bce0984cbc1b74320",
+    "git_blob_sha1": "0a2a7f524c1790dfaeb92d56eb76ec7c333d5c8e"
+  }
+]