EPICv2_thyroid_malignancy.clfx: pediatric thyroid nodule benign vs malignant, methscope random forest
Wanding Zhou
16816a6
A 500-tree methscope random forest over 3,000 single EPICv2 probes, trained
on 117 pediatric thyroid tumors (64 benign, 53 malignant; no BRAF V600E or
kinase fusion). Leave-one-patient-out it matches the lab's R forest: AUC
0.988 vs 0.986, 111/117 right at 0.5, Bethesda III+IV 55/58, DICER1 23/25.
SHA256SUMS adds the file; README's EPICv2 fetch line names it.
23 days ago
EPICv2 thyroid classifiers: invasiveness and driver group (CCR 2026), methscope random forests
Wanding Zhou
4baddd9
EPICv2_thyroid_invasiveness.clfx (High/Low) and EPICv2_thyroid_driver.clfx
(BRAF V600E, DICER1, Kinase Fusion, Ras-like): the pediatric thyroid
carcinoma classifiers of Li et al. (CCR 2026) rebuilt with methscope's
random forest (classify-train --framework rf) on the 87 / 84 reference-
cohort primaries. Leave-one-out they match the published R forests (85.1%
vs 83.9%, 92.9% vs 95.2%); on the validation primaries 76.1% vs 75.0% and
80.5% vs 84.1%. Scored by a methscope with the rf framework.
SHA256SUMS adds the two files; README adds the EPICv2 fetch line.
hg38_62celltypes, hg38_33celltypes: .msdref version 3, with a confusion
Wanding Zhou
ef6b77a
Both human deconvolution references now carry an MSDCONF1 trailer, which
`deconv --group-threshold` reads to report near-collinear classes under one
joined label rather than splitting mass between columns the data cannot
separate. The mouse reference has carried one since v11; these are the two
most readers fetch.
The M/U block of each is byte-identical to the version-2 artifact it
replaces. `cmp -l` over the old file's length reports exactly ONE differing
byte in each -- offset 9, the version word, 2 to 3 -- and the added bytes are
the appended trailer (20 + 4*n^2 + len(prov)): 15,624 for the 62-class,
4,584 for the 33-class. So this is not a rebuilt model that happens to agree;
it is the same model plus a trailer.
Measured on held-out data, because a shipped build uses everything and has
none left. The 62-class twin was rebuilt FROM CELLS -- pools over folds 1-9
of the Zhou 2025 hash split, fold 0 held out as mixture components, with
Hepatocyte and Granulocyte held out by donor sin…
mm10_41celltypes.msdref: the first mouse deconvolution reference
Wanding Zhou
43b7a27
A pure addition -- every v10 file ships unchanged, nothing withdrawn.
41 Liu 2021 mouse brain major types over the 6,765,325 CpGs that could ever
separate some class subset (30.9% of mm10), packed as uint16 M/U from
pseudobulks of all 54,447 cells with no fold held back. Until now every
.msdref was human and mouse shipped only classifiers plus the wg decoder.
It is the first version-3 artifact: it carries a validation confusion
trailer, which `deconv --group-threshold` reads to report near-collinear
classes under one joined label (CA3/DG-po) rather than splitting mass between
columns the data cannot separate. The trailer was measured on a
fold-held-out twin, since an all-cells build has no held-out cells to measure
with. A binary older than that format refuses the file rather than misreading
it, so the registry pin moves only with a YAME release that can read it.
Validated before staging: all 41 pools self-deconvolve to 100.0%; on P21
spatial DMT mouse brain (2,235 in-tissue pixels)…
SHA256SUMS: hg38_wg and both lites rehashed, two withdrawn files dropped
hg38_celltype is now a violation-framework model with no fitted parameters: it reads each class's expected methylation state off the MRMP binstrings and calls the class the query contradicts least, weighting patterns by sqrt(n_cpg). Confidence is the margin to the runner-up, which collapses when no class fits.
External test 0.842 / macro-F1 0.853 at >=100k covered CpGs. Colorectal cancer single cells call 100% gastrointestinal from liver and lymph metastases alike, and the sink that sent globally demethylated tumours to Acinar.(Pancreas) drops from 30.0% of TCGA calls to 2.3%.
Classifier bundles are renamed .ubjx -> .clfx: .ubjx named a payload format rather than a role and had already stopped being accurate (hg38_sex carried a plain-text linear spec, no UBJSON). Bundles are detected by MSBNDL1 magic, never by extension, so .ubjx stays accepted. The previous filenames remain available at --revision v5.
mm10_celltype.ubjx is unchanged, pending a Liu 2021 rebuild.