research stage
Current public records center on computational analysis and literature review. They are separated from clinical trials, animal experiments, and patient-application data.
RECOVER research records for understanding skin recovery more precisely
Research Ledger · Issue 01
Research record
RECOVER research records for understanding skin recovery more precisely
26 recordsdetailed public records
40 recordspublic DOI/OSF/Zenodo identifiers
14 targetsskin-recovery-related targets
50+ toolscomputational tool validation
770+references
research record principles
for understanding skin recovery more precisely
for understanding skin recovery more precisely
RECOVER research records
Indentation, hardening, persistent redness or pigmentation, and the speed of skin recovery cannot be explained by one phrase. RECOVER Research organizes literature, images, and computational research to read these changes more precisely. In clinic, this mindset helps separate scars by depth, tethering, color, and skin response.
research candidate compared computationally
computational analysis stage
An example research question at the computational-analysis stage.
- candidate library
EMB-3
R15
R16
R17
PVP-1
CHR-A
GENESIS MEDICINE LAB
research space
This RECOVER research space organizes literature, images, and computational studies by the same standard. Clinical scar reading is built on this background.
- Why we record
Skin recovery is difficult to explain simply as
Skin recovery is difficult to explain simply as
a wound closing.
The processes of indentation, hardening, persistent pigmentation, post-acne recovery speed, and collagen decline in aged skin involve cells, proteins, inflammation, the skin barrier, and individual response together.
RECOVER does not leave these questions to impressions alone. It reviews recovery-planning records, skin images, public literature, and computational analysis together.
The records on this page are questions for understanding skin recovery more precisely. Most are currently computational analysis and literature review records, shared as research background before human application.
research disclosure standards
- stage
- scope
- transparency
- language
scope of use
In clinic, this research is used as background and a standard for viewing scars.
transparent disclosure
Some research code and records are shared for external review.
communication standard
All language is limited to explaining the stage and source of research activity.
IChapter
Chapter 1 - research structure
How the research continues -
How the research continues -
reading, records, and computational research
Gangnam recovery planning record, capture· tool, computer-based research.
skin recovery record. Gangnam 5 · 2026 8 17 opening.
The technology company founded by Han Cheong Woo. It designs and develops the review screens and record tools that read 3D scan records directly, and operates them at RECOVER Gangnam.
Genesis_Medicine Lab
skin recovery research.
1 Genesis_Medicine Lab, Seoul, Republic of Korea
2 HAN PREDICT, Inc. (hanpredict.com)
3 Recover Korean Medicine Clinic (recover-clinic.kr)
ORCID · 0009-0004-4805-8815 ↗DStack
visit RECOVER research.
research·
research·
detailed records
Discovery Programs target, skin absorption, computational tool validation record public. RECOVER Research validation standard.
research record. DOI, OSF, Zenodo, computational tool, target record research validation standard check.
— 01 · Core Targets
Scar · ECM · MMP-1
We organize the five scar-regeneration targets (TGF-beta1, MMP-1, COL1A1, CTGF, LOX) alongside computational validation records for MMP-1-related structure panels.
record07
— 02 · Discovery
xtb · Boltz-2 · ADMET-AI
quantum chemistry, - cofold, drug-likeness, Bayesian active learning NNP 5-stack(MatterSim 5M, Orb-v3 OMol25, MACE-OMol25, eSEN, UMA), de novo design(RFdiffusion 3, LigandMPNN, FlowPacker, BoltzGen), pocket consensus(PocketMiner, DeepPocket, fpocket), NP-MS(SIRIUS+CFM-ID+MIST+DiffMS), TCM(TCMSP+BATMAN-TCM 2.0) validation.
record13
— 03 · Translation
Topical PBPK · Chronotherapy · Korean PGX
Dancik 4-layer skin logKp meridian chronotherapy(子午流注) chronotherapy, Korean PGX topical framework, Dongui Bogam·Hyangyak Jipseongbang scaffold xref candidate literature map.
record04
— 04 · Validation
ABFE · NNP · LigandMPNN · PoseBusters
We validate the limits and operating conditions of computational tools in public records with external identifiers. Reproducibility measures include three-NNP cross-agreement r=0.9142, PoseBusters v2 93.9% pass, and LigandMPNN Zn²⁺ recovery 95.3%.
record08
FCards
Featured Programs
organized for external sharing
organized for external sharing
three research candidate groups
Among 26 detailed public records and 35+ expanded records including paper_B v1 and paper #19 v2 outline, these candidate groups are prioritized for external review and collaboration inquiry. All candidates are treated as computational-analysis-stage research candidates.
S/N · 01-MMP1
— Card 01 · Journal Review Track
MMP-1
MMP-1
Zn²⁺ NNP cross-agreement
We examine conformer-energy agreement among three neural network potentials and coarse bound/unbound classification for MMP-1 Zn²⁺ active-site structure panels, without asserting compound identities or potency. Supporting technical reports (the OMol25 paradox, Boltz-2 steering, and de novo Zn design) document methodological limits and operating conditions externally.
target
in silico
deposit
published
peer-rev
subjectstructure panels · 3-NNP · Boltz-2 · ABFE failure mode
compositionpaper_A v6 (XXVI) + supporting tech reports XXIII · XXIV · XXV
evidence3-NNP Pearson r=0.9142 · MACE↔Orb r=0.9992 · PoseBusters v2 93.9% · σ 2,039× collapse
statusRevision in preparation
S/N · 02-EMB3
— Card 02 · Natural Product Lead
EMB-3
EMB-3
Embelia ribes scaffold
Ayurvedic · Korean herbal medicine Vidanga(Vidanga, Embelia ribes) embelin AI scaffold-hopping scars research candidate. computational analysis stage record.
target
in silico
deposit
wet-lab
IRB
targetMMP-1 · TGF-β1
modalityTopical small molecule
statusPreprint computational-analysis-stage record published
nextWet-lab pilot (fibroblast · TGF-β assay) collaboration
S/N · 03-R16R17
— Card 03 · Lead Optimization Stack
R16 / R17
R16 / R17
Chromanol optimization
6-cycle Bayesian active learning(R9–R14) chromanol multi-target binder. 14 skin-disease target robustness benchmark.
target
in silico
deposit
wet-lab
IRB
target14 skin-disease targets (multi-target)
modalityAI-generated chromanol series
statusPreprint 30 · 60 · 200 ns MD stability check
nextADMET-AI profiling · selectivity filter
— Publishable Findings · frontier metrics
— xtb · Boltz-2
— Reproducibility
— Structure validation
— De novo metal design
We disclose reproducibility numbers, not just the number of tools installed
These are cross-validation records that test whether conclusions hold when the same candidates and targets are filtered through multiple tools.
Outlier collapse
For the structure panel, σ_E collapses from 14.27 to 0.0069 kcal/mol after xtb-OPT pre-relaxation (2,039×). Boltz-2 confidence-axis outliers are reviewed separately.
Runtime · NNP agreement
Three-NNP conformer-energy cross-agreement is Pearson r=0.9142 [0.826, 0.971] (LOO ±0.0104; the lowest of the three pairs). MACE-OMol25 and Orb-v3 correlate at r=0.9992, disclosed as an effectively two-independent-engine comparison.
PoseBusters benchmark
We use the PoseBusters v2 93.9% pass rate as a structural plausibility check and record cofold-steering protocol failure modes separately.
LigandMPNN Zn²⁺ recovery
RFdiffusion 3 → LigandMPNN → FlowPacker → BoltzGen LigandMPNN Zn²⁺ recovery 95.3% vs ProtMPNN 46.4% paper_C public.
IIChapter
Chapter 2 - research methods
14 protein targets,
14 protein targets,
4 recovery questions
The 26 detailed records and 35+ expanded records compare molecular targets related to skin recovery computationally. Some tools and code are public for external reproducibility.
SCAR · 5
PIGMENTATION · 4
ALOPECIA · 2
ACNE · PHOTOAGING · 3
scar regeneration
TGF-β1MMP-1COL1A1CTGFLOX
pigmentation
TyrosinaseMITFTYRP1DCT
hair loss
SRD5A2AR
acne / photoaging
SREBP1PTGS2SIRT1
IIMethods
Chapter 2 - Computational Toolkit
50+ frontier-tech tools,
50+ frontier-tech tools,
validation matters more than installation
Alongside 12 representative tools, NNP 5-stack, de novo design, pocket consensus, NP-MS, TCM databases, and 60+ adapters cross-check the same questions across multiple computational axes.
Boltz-2
protein-candidate binding structure calculation
OpenMM 8
molecular dynamics simulation (~10-200 ns)
ADMET-AI
41 drug-likeness related metrics
RFdiffusion3 · LigandMPNN
Zn recovery 77.5% vs ProtMPNN 40.6% · FlowPacker/BoltzGen validation
NNP 5-stack
OMol25/OMat validation
xTB / GFN2
semi-empirical quantum chemistry (HOMO/LUMO)
RDKit
cheminformatics and SMILES handling
PoseBusters v2
93.9% pass rate used as a structural plausibility criterion
Open Targets 26.03
6-source independent corroboration
NP-MS · TCM stack
TCMSP · BATMAN-TCM 2.0 Korean herbal medicine scaffold xref
Dancik PBPK
four-layer skin absorption model
Frontier Stack
cross-validation results are prioritized over install logs
IIIRecords
Chapter 3 - Public Records
public research records,
public research records,
26 detailed records and 35+ expanded records
Academic records written and published by Genesis_Medicine Lab. The table keeps detailed records I-XXVI, while expanded records such as paper_B v1 and paper #19 v2 outline are reflected in the program narrative above. These are research-stage records, separate from clinical efficacy claims.
- Status badges: 26 detailed record stages
Preprint
12records
Tech Report
07records
Forthcoming
01planned
Records planned for deposit soon. They move to Preprint when an external identifier (DOI) is issued.
- Discovery Programs Navigator
— Program A
— Program B
— Program C
— Secondary Screens
I-XXVI detailed records and 35+ expanded records grouped by research program
The table keeps detailed records in Roman numeral order I-XXVI, while the upper chip strip highlights IX-XIII, XVIII, XIX, and the paper_B expansion axis. Records spanning two programs are shown in both.
Scar & ECM Remodeling
A research-candidate discovery stream for scar and ECM remodeling. It brings together the five targets (TGF-beta1, MMP-1, COL1A1, CTGF, LOX), multi-target chromanol series, and computational validation of MMP-1-related structure panels.
Topical Translation
skin absorption pharmacokinetics axis. Dancik 4-layer PBPK, meridian chronotherapy chronotherapy, Korean PGX topical framework.
Method Validation
computational tool validation axis. ZAFF-AMBER metal, OMol25 paradox, Boltz-2 steering, LigandMPNN, active learning multifidelity public record.
pigment · hair loss · acne · framework
vertical framework record. pigment · hair loss · acne / photoaging single-screen comparison PGx panel · Dongui Bogam atlas literature · framework.
— Program × Record · 4 × 26 matrix
columns I–XXVI · rows A · B · C · S
A · Scar/ECMscars · ECM
10
B · Topicalskin absorption
03
C · Methodtool validation
12
S · Secondaryauxiliary / framework
07
1234567891011121314151617181920212223242526
∑
A · scar/ECM
B · topical
C · method
S · secondary
cross-program (XV · XVI · XVII · XXI · XXVI)
— * cross-program · record
01
XXVI
Revision in preparation
Three-neural-network-potential cross-agreement and bound/unbound coarse classification for MMP-1 zinc active-site structure panels (without asserting identity or potency; revision in preparation)
Cross-Validation of Three Neural Network Potentials for MMP-1 Zn Active-Site Structure Panels: Reproducibility Agreement and Bound/Unbound Coarse Classification
We assess cross-model reproducibility over structure panels for the MMP-1 Zn²⁺ active site without asserting compound identities or potency. Three-NNP comparison yields a minimum pairwise Pearson r = 0.9142 (bootstrap 95% CI [0.826, 0.971]; leave-one-out ±0.0104); MACE-OMol25 and Orb-v3 are highly correlated (r = 0.9992) and therefore constitute effectively two independent engines. An upstream GFN2-xTB OPT pre-relaxation reduces the σ outlier from 14.27 to 0.0069 kcal/mol — a 2,039-fold reduction — and structure generation yields a 93.9% PoseBusters v2 pass rate. We report conformal calibration and coarse bound/unbound classification as reproducibility-focused computational assessments; no inhibitor potency or affinity ranking is claimed. Revision in preparation; the linked Zenodo v6 record has not yet been superseded. Sole author. In silico stage; no clinical efficacy claim.
02
XXIII
Tech Report
MMP-1 skin research computational tool validation record
The OMol25 Paradox: Training-Data Domain Specificity Outweighs Architecture in Neural Network Potential Performance on Zn²⁺ Metalloenzyme Conformer Energies
Machine-learning interatomic potentials (MLIPs) are widely assumed to dominate semi-empirical alternatives on molecular tasks. We test this assumption on the conformational energy landscape of fifteen Zn²⁺-binding matrix metalloproteinase-1 (MMP-1) inhibitors drawn from ChEMBL, generated by Boltz-2x cofold sampling with physicality steering. Across thirteen independent diffusion replicates, we evaluate ten energy functions on 100 conformers per ligand per replicate (253,500 conformer single points total). The headline finding is the OMol25 paradox: the same Orb-v3 architecture trained on Meta FAIR's OMol25 molecular dataset drops to r = 0.374 ± 0.025 against GFN1-xTB, versus r = 0.886 ± 0.009 for Orb-v3 OMat — a 0.512 Pearson gap with ≈ 68σ statistical significance across 13 replicates. The v5h revision (2026-05-15) adds a 17-cycle pipeline timing reproducibility table and a chain × concurrent-ADMET fair-share linear regression model. Cluster placement is set by training-data domain, not by architecture. We recommend that practitioners choose materials-trained or broadly-trained MLIPs over molecular-only-trained MLIPs for Zn²⁺ metalloenzyme conformer-ensemble work, pending higher-level DFT validation.
03
XXIV
Tech Report
- tool
Steering potentials, not bug fixes, eliminate catastrophic outliers in Boltz-2 cofold protein-ligand affinity prediction: a six-way protocol evaluation on zinc-hydroxamate-like MMP-1 active-site ligands
Boltz-2 cofold sampling for protein-ligand affinity prediction occasionally produces catastrophic outliers — single conformers with unphysical energies that distort downstream affinity ranking. We evaluate six protocol variants on the canary ligand CHEMBL94487 and the full MMP-1 hydroxamate cohort (15 inhibitors × 100 samples each), comparing standard Boltz-2 vs. community-fork bug-patched versions, with and without the --use_potentials physicality steering flag. The full-cohort outlier rate drops from 0.067% (standard Boltz-2 without flag) to 0.000% (standard Boltz-2x with --use_potentials, σ_filt = 4.29 kcal/mol). The community fork's metal-ion bug fix alone (without the flag) does not eliminate outliers and shows σ_filt = 6.66 kcal/mol. We recommend standard Boltz-2 + --use_potentials as the publishable protocol for Zn-metalloenzyme cofold work.
04
XXV
Tech Report
ECM/ public record
End-to-end de novo design of Zn²⁺ metallohydrolase binders: an open-source canonical pipeline anchored by LigandMPNN's metal-coordination recovery
De novo enzyme and binder design for Zn²⁺ metallohydrolases (MMP, HDAC, carbonic anhydrase families) requires a sequence-design step that respects the metal-coordination geometry of the active site. We benchmark the canonical four-stage open-source pipeline (RFdiffusion3 backbone generation → LigandMPNN sequence design → FlowPacker side-chain packing → AlphaFold3/Boltz-2x cofold validation) on 1HFC matrix metalloproteinase-1 catalytic domain. The headline finding: LigandMPNN doubles Zn-coordinating residue native recovery from 46.4% (plain ProteinMPNN) to 95.3%, with structural-Zn-triad recovery jumping from 0% to 90.6%. ESM-C 600M zero-shot pseudoperplexity confirms the LigandMPNN designs as more native-like (2.85 vs 3.03). The pipeline assembly and per-stage failure modes are documented for community reproduction.
05
VIII
Revision in preparation
natural product candidate computational
A calibrated absolute binding free energy pipeline with flat-bottom centroid restraint and analytical standard-state correction for natural-product scaffold-hopping in OpenMM 8 / openmmtools 0.26
This technical record documents absolute binding free energy (ABFE) pipeline components, including thermodynamic-cycle closure across complex- and solvent-decoupling legs, ligand restraints, and analytical standard-state correction. Its calibration and quantitative potency-assessment results are under revision; the linked Zenodo record has not yet been superseded.
06
XVIII
Tech Report
skin candidate computational research
Cost-Aware Multi-Fidelity Bayesian Optimization with Runtime-Gated Cascading Tiers for Autonomous in silico Dermatology Discovery
Autonomous virtual-screening pipelines suffer a recurring failure mode: cheap predictors (Boltz-2 affinity, ADMET, xTB conformer scoring) accumulate thousands of candidate scores, yet expensive validation tiers (long molecular dynamics, absolute binding free energy, wet-lab) advance unevenly because the scheduler cannot reason simultaneously about cost, information value, gate-policy, and runtime availability of each…
07
XIV
Tech Report
skin candidate computational
A Topical Skin PBPK Pipeline for Natural-Product-Inspired Therapeutics: Integrating Dancik 4-Layer ODE, SkinPiX-Trained LGBM logKp, and NPASS 2026 ADME Records
Topical (transdermal) drug delivery is the dominant route for dermatological therapeutics, yet most AI drug-discovery pipelines omit physiologically-based pharmacokinetic (PBPK) modeling for skin. We present a fully open-source pipeline combining: (1) Dancik 4-layer skin PBPK (stratum corneum → viable epidermis → dermis → systemic); (2) LightGBM logKp head trained on SkinPiX (n=2,326, OECD 428) + NPASS 2026 ADME reco…
08
XII
Tech Report
Genesis_Medicine — Korean herbal medicine research public AI
Genesis_Medicine: an open-source AI pipeline for Korean traditional medicine drug discovery — ~25 active modules + ~25 adapter scaffold catalog with design philosophy (v0.3 audit-corrected)
Modern AI-driven drug discovery is built on a heterogeneous stack of open-source tools. Specialized application domains require tool selection, integration, and adaptation. We describe Genesis_Medicine, an open-source pipeline for AI-driven Korean traditional medicine (Korean herbal medicine) drug discovery. We honestly distinguish: (i) a 7-tool active core that produces all real pipeline outputs in this work (Boltz-2 cofold, REINVEN…
09
XXII
Tech Report
natural product candidate computational records
Method-, conformer-count-, and solvent-robust ranking of natural-product screening corpora via xtb semi-empirical quantum chemistry: a multi-axis robustness benchmark
Tight-binding semi-empirical quantum chemistry (xtb-GFNn-xTB) has become a workhorse triage tool for ranking large compound libraries ahead of more expensive structure-based scoring. Whether the resulting ranks are robust to the methodological choices an analyst makes — the GFN parameter set, the conformer-ensemble size, and the implicit solvent model — is rarely tested at scale. We benchmark these three robustness a…
10
XXI
Revision in preparation
target computational
Limitations of ZAFF-AMBER absolute binding free energy calculation for zinc metalloenzyme inhibitors: a replicate-pair analysis on MMP-1
This technical record examines limitations of fixed-charge ZAFF-AMBER absolute binding free energy calculations for zinc metalloenzymes. Quantitative accuracy findings are under revision while a real-activity-based ABFE failure panel is being prepared; the linked OSF record remains under revision.
11
III
Preprint
EMB-3 skin scars research candidate
AI-driven scaffold-hopping of Embelia ribes embelin yields a topical-friendly anti-fibrotic candidate (EMB-3): an in silico case study for skin scar regeneration
Embelin (2,5-dihydroxy-3-undecyl-1,4-benzoquinone) is the principal bioactive of Embelia ribes Burm.f. (Ayurvedic Vidanga; East Asian Vidanga), an Ayurvedic and East Asian traditional-medicine plant with documented anti-fibrotic activity in liver and pulmonary models but no published investigation in skin fibrosis. We present an in silico case study in which embelin serves as the scaffold-hop seed for an AI-augmented …
12
I
Preprint
Vidanga skin research candidate
Embelia ribes (Vidanga, Vidanga) revisited: from Ayurvedic-East Asian traditional use to AI-augmented scaffold-hopping for skin fibrosis
Embelia ribes Burm.f. (Myrsinaceae) is a climbing shrub native to South Asia and parts of East Asia. Its dried fruit, known as Vidanga in Ayurvedic medicine and Vidanga(子團子) in Korean materia medica references, has been used for over two millennia as an anthelmintic, anti-inflammatory and digestive remedy. The principal bioactive constituent is embelin (2,5-dihydroxy-3-undecyl-1,4-benzoquinone), present at 4–5% of dry w…
13
IV
Preprint
pigment candidate Korean herbal medicine computational comparison
In silico screening of 15 Korean herbal compounds against tyrosinase, TYRP1 and DCT (TRP-2) for topical hyperpigmentation: real Boltz-2 cofold + ADMET-AI results identify oxyresveratrol, curcumin, and resveratrol as topical-friendly leads
Topical hyperpigmentation disorders (melasma, post-inflammatory hyperpigmentation, solar lentigines) are mediated by the tyrosinase (TYR) / TYRP1 / DCT (TRP-2) melanin-synthesis network and the master transcription factor MITF. Korean traditional medicine maintains a long-standing repertoire of (depigmenting) preparations centered on Glycyrrhiza uralensis (licorice), Camellia sinensis (green tea), Morus alba (mulberry root bark), *B…
14
XV
Preprint
skin target natural product candidate research
A Universal Pterocarpan-Vinyl-Polyphenol Scaffold for Multi-Target Skin Therapeutics: Six-Cycle Bayesian Active Learning Identifies Six Multi-Target Leaders Across 14 Skin-Disease Targets
We report the discovery of a pterocarpan-vinyl-polyphenol scaffold family that acts as a universal molecular template across six independent skin-disease verticals (scar regeneration, pigmentation, alopecia, acne, photoaging, fibrosis cross-disease). Six members of this family were identified through six rounds of Bayesian active learning (R9–R14, 4,597 cofold predictions integrated, 14 protein targets, 200-candidate…
15
XVII
Preprint
candidate computational
R16 Topical Chromanol Lead Short Communication: 18-Pair 30 ns Matrix, 60 ns Robustness Panels, and Complete 200 ns Anchor-Triad Follow-up
R16 optimized the R15 chromanol fragment toward a topical lead hypothesis by increasing skin-window compatibility while preserving a compact chromanol core. We evaluated six chloro/dimethyl analogs across TGFB1, DCT, and TYR using Boltz-2 cofolding, an 18-pair 30 ns OpenMM stability matrix, two 60 ns robustness panels, and 100 ns plus 200 ns anchor-triad follow-up. The top cofold row was r16_03_tgfb1 (R15_chromanol_C…
16
XVI
Preprint
candidate
R15 Chromanol Safety-First Fragment Triage: 14-Target Cofolding, ADMET/xTB Filtering, and 30 ns MD Separates Systemic-Safety and Topical-Lead Paths
R15 generated a compact chromanol fragment, OCC1COc2cc(O)ccc2C1, as a safety-first derivative of the broader pterocarpan-vinyl-polyphenol scaffold program. The central question is whether this fragment should be treated as a topical lead or as a systemic-safety fragment hypothesis. The answer is deliberately split. In ADMET/xTB triage, the R15 parent is predicted to be clean for AMES, DILI, and hERG liability, but it…
17
XX
In Prep
RECOVER R17: candidate map
RECOVER R17: chromanol generative atlas (in silico)
Manuscript metadata to be finalized. DOI issued; full abstract pending publication.
18
V
Preprint
hair loss candidate Korean herbal medicine computational comparison
In silico screening of 15 Korean herbal compounds against the SRD5A2 / AR / β-catenin axis for androgenetic alopecia: real Boltz-2 data identifies Emodin, Saponin Re, and Biochanin A as topical-scalp candidates with safety considerations
Androgenetic alopecia (AGA) is driven by androgen-pathway enzymes (SRD5A1/2 catalyzing testosterone → DHT), the androgen receptor (AR), and the hair-follicle-cycle Wnt / β-catenin axis. Korean traditional medicine documents hair-vitalization preparations centered on Polygonum multiflorum (Polygonum multiflorum; emodin, physcion), Astragalus membranaceus (Astragalus; astragaloside IV), and Panax ginseng (ginseng; ginsenosides Rg1, Rb1, Saponi…
19
XIII
Preprint
hair loss research
Mechanotransduction in Androgenetic Alopecia: An In Silico Repositioning Study of PIEZO1 + MLCK Axis Using Cofolding and Pilosebaceous Single-Cell Atlas Constraints
Androgenetic alopecia (AGA) has been treated for decades through the androgen axis (5α-reductase / AR), yet ~50% of patients respond inadequately. A 2026 Nature Communications report identified connective tissue sheath (CTS) hypercontractility mediated by PIEZO1 mechanosensation and myosin-light-chain kinase (MLCK) as a non-androgen cause of follicular miniaturization, with ML-7 (MLCK inhibitor) restoring hair grow…
20
VII
Preprint
candidate comparison
In silico screening of 15 polyphenols and reference compounds against MMP-1 + SIRT1 for topical photoaging: real Boltz-2 data positions EMB-3 at the top of the panel; Resveratrol leads SIRT1 axis; classical references (vitamin C, niacinamide, ferulic acid) rank low
Photoaging — UV-induced premature dermal change including wrinkle formation, elastosis, pigment irregularity — is mediated by MMP-1 (interstitial collagenase, the principal photoaging effector) and SIRT1 (NAD-dependent deacetylase, longevity / DNA-repair regulator), among other targets. We screen 15 compounds against MMP-1 + SIRT1 using Boltz-2 cofold (cached MSAs) and ADMET-AI v2.0.1. FBN1, mTOR, and elastin were NO…
21
VI
Preprint
acne candidate Korean herbal medicine computational comparison
In silico screening of Korean herbal compounds against the SRD5A2 / AR sebaceous-androgen axis for inflammatory acne: real Boltz-2 data identifies Baicalein as top topical-friendly candidate; Berberine carries a critical hERG safety flag
Inflammatory acne pathogenesis includes androgen-driven sebaceous activity (SRD5A1/2 + AR + SREBP1), ductal hyperkeratinization, Cutibacterium acnes colonization with virulence factor expression (RoxP, GehA, sortase), and inflammatory cascades. Korean traditional medicine documents anti-acne preparations centered on Coptis (Coptis chinensis; berberine, palmatine), Scutellaria (Scutellaria baicalensis; baicalein, baicalin, w…
22
IX
Preprint
skin scars research research
From skin scar to systemic fibrosis: Open Targets evidence and the limits of canonical anti-fibrotic axis-based cross-disease hypothesis (an EMB-3 in silico case)
Pathological fibrosis across organs — skin scarring, idiopathic pulmonary fibrosis (IPF), systemic sclerosis, renal fibrosis, hepatic fibrosis — is widely framed in medicinal-chemistry literature as sharing a converging TGF-β / Smad / MMP / CTGF / collagen-deposition master-switch network. We investigated whether this conceptual axis-sharing translates to evidence-based cross-disease applicability for an AI-derived m…
23
XIX
Preprint
paper #19 v1 · Dongui Bogam·Hyangyak Jipseongbang herbal medicine target public scaffold atlas
Korean herbal medicine scaffold cross-reference: a literature-grounded scaffold atlas linking Dongui Bogam and Hyangyak Jipseongbang monographs to modern target classes via 60 ns molecular dynamics evidence
Korean traditional medicine texts (Dongui Bogam 1610, Hyangyak Jipseongbang 1433) catalog hundreds of herbal monographs with empirical clinical use across centuries, but systematic cross-reference between these monographs and contemporary molecular target classes is fragmentary. We assemble a scaffold-level atlas mapping Dongui Bogam / Hyangyak Jipseongbang monograph entries to modern phytochemistry and target classes via 60 ns molecular dynamics evidence. Hypothesis-generating only; no clinical efficacy claim.
24
XI
Framework
Korean herbal medicine
A Korean pharmacogenomics panel for topical-herb-medicine personalization: framework design integrating CYP, UGT, HLA, and skin-barrier variants
Pharmacogenomic (PGx) variation drives substantial inter-individual differences in drug response. The Korean population exhibits distinctive allele frequencies relative to the Caucasian-dominated reference data underlying many published PGx panels — for example, CYP2D6\10 (intermediate metabolizer, ~50% frequency in Koreans vs ~3% Caucasian [1]) and HLA-B\15:02 (severe cutaneous adverse-reaction risk, frequency ~0.…
25
X
Framework
research
Chronotherapy of topical Korean medicine: a quantitative framework integrating meridian chronotherapy(子午流注) twelve-meridian timing with circadian molecular biology and topical pharmacokinetics
Korean traditional medicine maintains a centuries-old chronotherapeutic framework — meridian chronotherapy(子午流注, Cha-Oh-Ryu-Ju) twelve-meridian timing — in which the twelve principal meridians (///////////) are assigned 2-hour windows of peak qi flow (time window, sì-jin) across the 24-hour cycle. Modern circadian biology has revealed substantial molecular evidence for cell-autonomous and tissue-level circadian rhythms, includ…
26
II
Framework
Korean medicine clinic AI record tool research
An integrated AI workflow for a Korean medicine clinic: 3D facial diagnostics, RAG-based electronic medical records, and computational molecular prescription
Korean medicine (Korean medicine) clinics face a structural gap between traditional pattern-based diagnosis and modern personalized molecular medicine. We describe an integrated end-to-end workflow assembled across three affiliated entities — HAN PREDICT, Inc. (AI healthcare technology platform), Genesis_Medicine Lab (in silico natural-product drug discovery), and Recover Korean Medicine Clinic (skin-regeneration practice o…
26 detailed public records and 35+ including expansions/outlines; all records include external identifiers — Genesis_Medicine Lab · 2026
CInquiry
Collaboration Inquiry
For wet-lab validation and methodology collaboration,
For wet-lab validation and methodology collaboration,
collaboration proposals are welcome
Genesis_Medicine Lab research in silico stage. The next stage requires wet-lab validation by external labs, domain database collaboration, and joint computational methodology development.
Fibroblast · TGF-β · MMP-1 inhibition · ECM remodeling assay
skin R&D · research fibroblast / TGF-β / MMP-1 inhibition / ECM assay lab research candidate validation.
herbal medicine, natural product DB, and computational annotation
Chemical structure annotation and traditional medicine cross-reference data can be organized with herbal medicine and natural product DB operators such as KMCRIC, NPAtlas, and COCONUT.
joint AI-methodology development with Boltz, MACE, Orb, SevenNet, UMA, and related tools
This work jointly validates operating conditions and limits of AI computational-chemistry tools such as Boltz, MACE, Orb, SevenNet, and UMA, then publishes records with external identifiers.
— Direct contact
crazat7@gmail.com
Beyond these three areas, in silico external review and discussion inquiries are welcome. Replies are usually sent within 2-3 business days. Inquiries are limited to research activity, not medical advertising.