Multi-target evaluation of Korean herbal compounds against the tyrosinase / TYRP1 / DCT axis for topical hyperpigmentation: Boltz-2 cofold, ChEMBL-anchored ranking, ADMET-AI 107-endpoint safety panel, and 30 ns molecular dynamics validation of oxyresveratrol and curcumin leads

Correction (new version, 2026-07-21) — citation correction. Correction (2026-07-21) - calibration-anchor attribution and metric. Background: two calibration anchors had been conflated across this corpus. The 'Spearman |rho| ~ 0.72' anchor (n=15) was computed against the ligand panel found to be fabricated (2026-07-16 audit) and is void. The separate 'Pearson R = -0.453, n=93' anchor is REAL - it comes from the ChEMBL-derived 93-compound cohort of companion preprint #18 (pilot/cpu_meaningful/chembl_boltz2_calibration.csv, carrying per-record ChEMBL document identifiers) and reproduces on recomputation (Pearson -0.4535, Spearman -0.4582). It was, however, being cited incorrectly. In this record the abstract attributed R = -0.453 to 'the parallel MMP-1 calibration set', which reads as the fabricated panel and is wrong on two counts: the figure comes from preprint #18's real cohort, not from this project's withdrawn panel and not from preprint #8 (which never reported it); and it was measured on Boltz-2 affinity_pred_value, whereas this screen ranks on affinity_prob_binary. On the same 93 compounds affinity_prob_binary correlates with pIC50 at Pearson +0.178 / Spearman +0.048 - essentially no potency signal - so the anchor never certified the metric used here. The abstract now says so and the rankings are presented as relative and uncalibrated. No result, ranking or figure value changes; this is a provenance and scope-of-claim correction. Full audit: preprints/_metadata/DANGLING_CITATION_AUDIT_2026_07_21.md; underlying finding: preprints/_metadata/FABRICATED_PANEL_SCOPE_2026_07_16.md. Topical hyperpigmentation disorders (melasma, post-inflammatory hyperpigmentation, solar lentigines) are mediated by the melanin-synthesis network of tyrosinase (TYR), tyrosinase-related protein 1 (TYRP1), and dopachrome tautomerase (DCT, also known as TRP-2), with downstream regulation by the master transcription factor MITF. Korean traditional medicine maintains a long-standing repertoire of depigmenting preparations centered on Glycyrrhiza uralensis (Gamcho), Camellia sinensis (Nokcha), Morus alba (Sangbaekpi), and Broussonetia kazinoki (Daknamu). We evaluate a 15-compound curated Korean herbal library against the TYR / TYRP1 / DCT three-target panel using a four-layer in silico framework: (i) Boltz-2 protein-ligand co-folding (n = 45, cached MMseqs2 multiple-sequence alignments, RTX 5090 GPU); (ii) a ChEMBL pIC50 calibration anchor carried over from the parallel MMP-1 calibration set (Pearson R = -0.453, n = 93), establishing that Boltz-2 affinity_probability_binary is a relative-ranking predictor and not an absolute potency estimate; (iii) 30 ns OpenMM / GAFF-2.11 / TIP3P molecular dynamics on the top three mean-affinity candidates against TYR and DCT (six MD simulations, all RMSD below 2.0 Angstrom in the last-third trajectory window, demonstrating dynamic stability of the predicted binding poses); and (iv) ADMET-AI v2.0.1 safety prediction across 107 endpoints including hERG, skin reaction, AMES, and ClinTox. MITF was not folded due to absence of a cached MSA and is an explicit limitation of the present evaluation. Boltz-2 results identify oxyresveratrol from Morus alba root as the top mean-affinity candidate (mean 0.758, MD-stable across TYR and DCT) but with notable predicted skin-irritation and AMES liabilities that we do not minimize. Curcumin from Curcuma longa emerges as the cleanest topical-friendly alternative (mean 0.695, logP 3.29, hERG 0.07, MD-stable). Glabridin and licochalcone A from Glycyrrhiza uralensis score high on affinity but fall outside the topical absorption sweet spot (logP 3.99 and 4.46). Classical depigmenting references including kojic acid, arbutin, hydroquinone, and niacinamide rank low (mean 0.32 to 0.50), consistent with the Boltz-2 classifier's reduced sensitivity for fragment-size compounds and confirming the calibration limitation of the relative-ranking interpretation. Open Targets evidence-tier mapping places TYR, TYRP1, and DCT in the green-evidence tier for hyperpigmentation indications, supporting the target choice. We position oxyresveratrol and curcumin as topical-formulation candidates whose forward path is B16F10 melanin-content assay and a three-dimensional reconstructed-skin pigmentation model. All findings are in silico; no experimental potency, no clinical claim, and no commercial product is reported. Code, raw cofold JSON, MD trajectories, and ADMET predictions are released under Apache-2.0. Keywords: hyperpigmentation, melasma, tyrosinase, TYRP1, DCT, Korean traditional medicine, multi-target in silico evaluation, Boltz-2 cofold, ChEMBL calibration anchor, OpenMM molecular dynamics, ADMET-AI, oxyresveratrol, curcumin. ---