Molecular Arrest in Biological Regulation

Hypothèse exploratoire et framework pilote pour l'étude des composés à effets d'amortissement biologique

📢 Version actuelle : v1.1.2 | 📥 Télécharger la dernière release | 📖 Notes de version

---

Version actuelle (latest): v1.1.2
Date de release : 14 novembre 2025
DOI Zenodo : 10.5281/zenodo.17420685
Note: Extended dataset release with 25 candidate molecules (15 with complete/partial data)

Author: Tommy Lepesteur
ORCID: 0009-0009-0577-9563
License: CC-BY 4.0 (data), MIT License (code)
Repository: https://github.com/Mythmaker28/arrest-molecules

ℹ️ Cette page pointe toujours vers les assets de la dernière release. Les liens ci-dessous se mettent à jour automatiquement.

---

⚠️ Avertissement Méthodologique

Ce projet présente une hypothèse de travail exploratoire, non un cadre établi.

• ✅ Points forts: Transparence des données, code reproductible, prédictions falsifiables, classement par confiance
• ⚠️ Limitations critiques:
  • Core dataset: 10 composés validés (échantillon pilote, reproductibilité garantie)
  • Extended dataset (v1.1.2): 25 composés candidats (15 avec données complètes/partielles, 10 avec placeholders NR/EST)
  • Aucune donnée expérimentale nouvelle générée
  • Métriques proposées (API, EMC, NCR, etc.) ne sont pas encore validées empiriquement
  • 59% des prédictions sont de confiance modérée ou faible
  • Classifications basées sur des seuils arbitraires nécessitant validation

Ce travail vise à stimuler la recherche, pas à fournir des conclusions définitives.

---

Description

This data package accompanies the manuscript "Molecular Arrest in Biological Regulation: A Working Hypothesis for Natural Compounds with Dampening Effects" submitted to Frontiers in Pharmacology.

⚠️ IMPORTANT: This is a pilot hypothesis-generating framework, not an established paradigm. The work proposes testable concepts requiring prospective experimental validation.

The package contains:

• Core validated dataset: 10 exemplar compounds with complete data: 8 arrest agents (salvinorin A, paclitaxel, rapamycin, capsaicin, tetrodotoxin, resveratrol, ibogaine, noribogaine) + 2 oscillatory controls (psilocybin, LSD)
• 🆕 Extended candidate dataset (v1.1.2): 25 additional molecules across 6 pharmacological classes:
  • KOR agonists (nalfurafine, mesyl salvinorin B, U50,488, U69,593, enadoline)
  • mTOR inhibitors (everolimus, temsirolimus, ridaforolimus, zotarolimus, biolimus A9)
  • GABA-A modulators (muscimol, diazepam, midazolam, zolpidem, propofol, etomidate, thiopental)
  • Adenosine A1 agonists (adenosine, CPA, CCPA)
  • Negative controls (curcumin, quercetin, EGCG)
  • α2-adrenergic (dexmedetomidine)
• Data Quality Framework: Explicit Tier A/B/C/D classification with completeness metrics (see Data_Package_FAIR2/DATA_QUALITY_OVERVIEW.md)
• Proposed pharmacological metrics (API, EMC, NCR, AKR, PARI) — indices requiring empirical validation
• Uncertainty quantification via Monte Carlo simulation
• Confidence grading for 44 quantitative predictions (41% high confidence, 30% moderate, 30% low)
• Executable code for reproducing all calculations
• 5 extended case studies (ibogaine/noribogaine, resveratrol, fasting/breathing, psilocybin/LSD, AI memory)
• Data dictionary and usage protocols

⚠️ LIMITATIONS:

• No new experimental data were generated. All values are literature-derived (100+ sources).
• Core dataset: 10 compounds (validated, Tier A/B) — conclusions remain tentative, reproductibility guaranteed
• Extended dataset (v1.1.2): 25 compounds with varying data completeness:
  • 15 with complete or substantial data (Tier B/C: 60-90% completeness)
  • 10 with partial data requiring literature extraction (Tier D: <50% completeness)
  • Extended CSV not used by default in validation scripts (preserves core reproducibility)
• Metrics are proposed, not validated: API, EMC, NCR, AKR, PARI require prospective testing.
• Specific nuance for Salvinorin A: KOR pharmacology evidence is VERY HIGH, but the Level 3 arrest classification is a hypothesis; imaging evidence (fMRI/PET) exists in humans/primates yet remains limited to a few small studies, not a large, systematic program.
• Psychedelic comparators (psilocybin, LSD, ayahuasca/DMT): These are oscillatory/high-entropy comparators, NOT arrest molecules. They disrupt DMN connectivity and increase network entropy (EMC > 0), providing a contrast to the arrest candidates (EMC < 0).
• Many predictions are low-to-moderate confidence: See Confidence_Grading_Matrix.csv and DATA_QUALITY_OVERVIEW.md for details.

---

Files Included

Core Data Tables

1. Compound_Properties_Database.csv (10 rows × 36 columns) ✅ VALIDATED

• Molecular properties: formula, MW, logP, rotatable bonds, SMILES, InChI
• Binding parameters: K_i, K_d, EC₅₀, k_off
• Pharmacokinetics: t₁/₂, C_max, AUC, V_d, clearance, protein binding
• Compounds: Salvinorin A, Paclitaxel, Rapamycin, Capsaicin, Tetrodotoxin, Resveratrol, Ibogaine, Noribogaine, Psilocybin, LSD
• Literature sources: PubMed IDs for each parameter
• Status: Stable, reproducible, passes all validation tests

1.1 Compound_Properties_Experimental_Extended.csv (25 rows × 36 columns) 🆕 EXTENDED (v1.1.2)

• Same structure as core dataset but with varying data completeness
• Compounds (25 total):
  • KOR agonists (5): Nalfurafine, Mesyl Salvinorin B, U50,488, U69,593, Enadoline
  • mTOR inhibitors (5): Everolimus, Temsirolimus, Ridaforolimus, Zotarolimus, Biolimus A9
  • GABA-A modulators (7): Muscimol, Diazepam, Midazolam, Zolpidem, Propofol, Etomidate, Thiopental
  • A1 agonists (3): Adenosine, CPA, CCPA
  • Negative controls (3): Curcumin, Quercetin, EGCG
  • α2-adrenergic (1): Dexmedetomidine
• Purpose: Framework extension, SAR validation, class comparisons, negative controls
• Data Quality: Variable (Tier B/C/D) — 15 compounds with substantial data, 10 requiring extraction
• Completeness: 15-100% depending on compound (explicit NR/NA/ND/EST placeholders)
• Confidence: 2 HIGH, 5 MODERATE-HIGH, 6 MODERATE, 2 LOW (10 compounds pending full extraction)
• Usage: ⚠️ Not used by default validation scripts (preserves core reproducibility)
• See Data_Package_FAIR2/CANDIDATE_MOLECULES_TODO.md for extraction status and Data_Dictionary.md section 1.1

1.2 DATA_QUALITY_OVERVIEW.md 🆕 Quality Framework

• Explicit Tier A/B/C/D classification for core + extended datasets
• Completeness metrics and confidence justifications
• Usage guidelines (which tiers suitable for what analyses)
• Roadmap for compound promotion and data extraction priorities

2. API_Calculations_Full.xlsx (multi-sheet workbook)

• Sheet 1: Input parameters with literature sources
• Sheet 2: Step-by-step API calculations (absolute → relative)
• Sheet 3: Monte Carlo simulation results (10,000 iterations per compound)
• Sheet 4: 95% confidence intervals
• Sheet 5: Sensitivity analysis (varying parameters ±30%)

3. Confidence_Grading_Matrix.csv (44 rows × 6 columns)

• All quantitative predictions from manuscript
• Evidence type (direct/indirect/extrapolated)
• Confidence level (high/moderate/low)
• Validation requirements

4. Experimental_Protocols_Summary.csv (3 rows × 12 columns)

• Design parameters for Experiments 1-3
• Sample sizes with power calculations
• Primary/secondary outcomes
• Success/falsification criteria
• Estimated costs and timelines

Code and Scripts

5. Python_Code_API_Monte_Carlo.py

• Fully commented Python 3.8+ script
• Calculates API with uncertainty propagation
• Requires: numpy, pandas, matplotlib
• Runtime: <10 seconds on standard laptop
• Outputs: API values with 95% CI, diagnostic plots

6. R_Code_Figures_S2.R

• Generates 3-panel oscillatory advantage figure
• Requires: ggplot2, dplyr, survival, patchwork
• Customizable parameters (colors, font sizes)
• Exports 300 dpi TIFF files

Documentation

7. Data_Dictionary.md

• Complete variable definitions
• Units and measurement methods
• Abbreviations and ontology terms
• Quality control procedures

8. Literature_Search_Strategy.md

• PubMed search terms and filters
• PRISMA-style flowchart (1,247 abstracts screened → 95 retained)
• Inclusion/exclusion criteria
• Data extraction protocol

9. Case_Studies_Supplement.md ⭐ NEW

• Extended Case Study 1: Ibogaine & Noribogaine (hybrid arrest, addiction reset)
• Extended Case Study 2: Resveratrol & SIRT1 (minimal arrest, negative control)
• Extended Case Study 3: Fasting & Breathing (natural oscillators)
• Extended Case Study 4: Psilocybin & LSD (high-entropy oscillation, DMN dissolution)
• Extended Case Study 5: AI Memory (computational extension of arrest principles)

---

⚡ Quick Start (60 seconds)

# 1. Clone & install
git clone https://github.com/Mythmaker28/arrest-molecules.git
cd arrest-molecules
python -m venv .venv

# 2. Activate environment
# Windows: .venv\Scripts\activate
# Linux/Mac: source .venv/bin/activate

# 3. Install dependencies
pip install -r requirements.txt

# 4. Run reproducibility pipeline
python run_arrest_pipeline.py

Output: Validates data files, creates outputs/ directory, and runs analysis scripts if available.

📖 Detailed guide: See QUICKSTART.md and DOCS/REPRO_STATUS.md

---

Quick Start Guide

For users wanting to verify API calculations:

1. Open Compound_Properties_Database.csv
2. Identify compound of interest (e.g., Rapamycin)
3. Note parameters: K_d = 0.1 nM, τ_residence = 120 min, t_onset = 1440 min, EC₅₀ = 1 nM
4. Run Python script:

   python Python_Code_API_Monte_Carlo.py --compound Rapamycin

5. Output: API = 0.12 [95% CI: 0.08-0.16], Confidence: MODERATE

For users wanting to test the framework with new compounds:

1. Gather required parameters from literature (K_d, k_off or duration, t_onset, EC₅₀)
2. Add row to Compound_Properties_Database.csv
3. Run Python script with --new_compound flag
4. Compare API to reference standards (Table 1 in manuscript)
5. Critically evaluate whether arrest classification criteria (EMC/NCR/PARI) apply

Caution: The classification system is a working hypothesis. Empirical validation against experimental outcomes is essential before drawing conclusions.

For users wanting to reproduce figures:

1. Ensure R packages installed: install.packages(c("ggplot2", "dplyr", "patchwork"))
2. Run: Rscript R_Code_Figures_S2.R
3. Figures saved to ./output/ directory as TIFF (300 dpi)

---

Data Dictionary (Abbreviated)

Compound_Properties_Database.csv

Column Name	Description	Units	Data Type	Example
Compound_Name	Chemical name	—	String	Salvinorin A
CAS_Number	Chemical Abstracts Service registry	—	String	83729-01-5
SMILES	Simplified molecular-input line-entry system	—	String	COC(=O)[C@]12[C@@]3...
InChI	International Chemical Identifier	—	String	InChI=1S/C23H28O8...
Molecular_Formula	Elemental composition	—	String	C23H28O8
Molecular_Weight	Molecular mass	g/mol	Numeric	432.47
LogP	Octanol-water partition coefficient	—	Numeric	2.73
Rotatable_Bonds	Count of freely rotatable bonds	—	Integer	3
Primary_Target	Main molecular target	—	String	Kappa-opioid receptor
Target_Gene	Gene symbol	—	String	OPRK1
K_i	Inhibition constant	nM	Numeric	1.8
K_i_Source_PMID	PubMed ID for K_i value	—	Integer	12202542
K_d	Dissociation constant	nM	Numeric	1.8
k_off	Dissociation rate constant	min⁻¹	Numeric	0.04
tau_residence	Residence time (1/k_off)	min	Numeric	25
t_onset	Time to 50% effect	min	Numeric	1
EC50	Half-maximal effective conc.	nM	Numeric	2
EC50_Assay	Functional assay type	—	String	GIRK activation
t_half_plasma	Plasma half-life	h	Numeric	0.15
Cmax	Peak plasma concentration	ng/mL	Numeric	2.4
AUC	Area under curve	ng·h/mL	Numeric	15
Vd	Volume of distribution	L/kg	Numeric	3.2
Clearance	Systemic clearance	L/h/kg	Numeric	12.5
Protein_Binding	Plasma protein binding	%	Numeric	89
API_absolute	Arrest Potency Index (absolute)	nM⁻²	Numeric	6.95
API_relative	API normalized to salvinorin A	—	Numeric	1.00
API_CI_lower	95% CI lower bound	—	Numeric	0.85
API_CI_upper	95% CI upper bound	—	Numeric	1.15
AKR	Arrest Kinetics Ratio	—	Numeric	1.5
EMC	Entropy Modulation Coefficient	—	Numeric	-0.4
NCR	Network Connectivity Reduction	%	Numeric	50
PARI	Post-Arrest Resilience Index	—	Numeric	0.3
Arrest_Level	Classification (1/2/3)	—	String	Level 3
Confidence_Grade	Overall data quality	—	String	MODERATE

Missing Data Encoding

• NA: Not applicable (e.g., EMC for non-neural compounds)
• ND: Not determined (measurement not yet performed)
• NR: Not reported in literature
• EST: Estimated value (not directly measured)

---

Usage Notes

Target Audience

• Pharmacologists interested in testing the framework's predictions
• Medicinal chemists exploring potential dampening agent design principles
• Systems biologists studying network dynamics and oscillatory processes
• Researchers evaluating chronopharmacology hypotheses
• Educators seeking quantitative pharmacology case studies

Note: This framework is exploratory. Users should critically evaluate claims and contribute to validation efforts.

How to Cite This Work

For the dataset:

Lepesteur T. (2025). Molecular Arrest Framework Research Data Package (v1.1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.17420685

For the manuscript:

Lepesteur T. Molecular Arrest in Biological Regulation: A Unifying Framework for Natural Compounds with Dampening Effects. Manuscript in preparation (2025)

For the code:

Lepesteur T. (2025). molecular-arrest-framework: API calculation tools (v1.1.0). GitHub. https://github.com/Mythmaker28/arrest-molecules

BibTeX:

@dataset{lepesteur2025molecular,
  author       = {Lepesteur, Tommy},
  title        = {Molecular Arrest Framework Research Data Package},
  year         = 2025,
  publisher    = {Zenodo},
  version      = {v1.1.0},
  doi          = {10.5281/zenodo.17420685},
  url          = {https://doi.org/10.5281/zenodo.17420685}
}

Note: This DOI is the concept DOI that always points to the latest version. For citing a specific version, use the version-specific DOI from the Zenodo record.

---

Version History

v1.1.1 (October 2025): ⭐ CURRENT (Reproducibility Patch)

• ✅ Quick check script for instant validation (< 1s)
• ✅ Full CI/CD reproducibility workflow
• ✅ Automated artifact verification
• ✅ SHA256 checksums for all assets
• ✅ DOI Zenodo integrated everywhere
• ✅ 1-click reproducibility guaranteed (< 3 min)

v1.1.0 (October 2025):

• Extended dataset: 6 → 10 compounds (+67%)
• New compounds: Ibogaine, Noribogaine (hybrid arrest), Psilocybin, LSD (oscillation)
• Case studies supplement: 5 detailed case studies spanning arrest-oscillation continuum
• requirements.txt: Explicit Python dependencies with versions
• Updated predictions: 42 → 44 with refined confidence grading
• Enhanced documentation: 95+ literature sources

v1.0 (October 2025):

• Initial release accompanying manuscript submission
• 6 core arrest compounds characterized
• 44 predictions with confidence grading
• Monte Carlo uncertainty quantification implemented

Planned updates (contingent on validation):

• v1.2: Add salvinorin A analogs (8 compounds) from Supplementary Table S2 — if SAR patterns hold
• v2.0: Incorporate Experiment 1 results (salvinorin fMRI data) — if predictions confirmed
• v2.1: Incorporate Experiment 2 results (oscillatory cellular lifespan) — if oscillatory advantage validated
• v3.0: Clinical validation from Experiment 3 (TRD trial) — if safety/efficacy demonstrated

Note: Future versions depend on prospective experimental validation. If key predictions fail, the framework will require substantial revision or abandonment.

---

Contact and Support

Questions: tommy.lepesteur@hotmail.fr
Issue tracking: https://github.com/Mythmaker28/arrest-molecules/issues
Contributions: Pull requests welcome for novel compound additions (requires literature sources)

Controlled access requests (for synthesis protocols): Email corresponding author with:

1. Institutional email (no personal addresses)
2. IRB approval documentation (PDF)
3. Research protocol summary (1 page)
4. Statement of intended use (therapeutic research only)

Response within 7 business days.

---

License and Reuse

Data: Creative Commons Attribution 4.0 International (CC-BY 4.0)

• ✓ Share and adapt freely
• ✓ Provide attribution
• ✓ Indicate if changes made

Code: MIT License

• ✓ Use commercially or non-commercially
• ✓ Modify and distribute
• ✓ Include original license notice

Restrictions: Synthesis protocols for high-potency analogs subject to controlled access (see Data Availability Statement in manuscript Section "Data Availability Statement").

---

Acknowledgments

Data compilation supported by independent literature review with quality verification by an external consultant. Database access via freely available public resources (DrugBank, PubChem, ChEMBL).

---

Changelog & Release Notes

v1.1 (2025-10-21) - Extended Dataset & Case Studies ⭐

Dataset expansion (+67%) :

• 4 nouveaux composés ajoutés : Ibogaine, Noribogaine, Psilocybin, LSD
• Dataset : 6 → 10 composés couvrant tout le continuum arrest-oscillation
• Ibogaine/Noribogaine : Arrest hybride DAT/SERT/κ-opioid, mécanisme addiction reset
• Psilocybin/LSD : Oscillation haute entropie (EMC positif), dissolution DMN

Nouveau supplément : Case_Studies_Supplement.md (5 études détaillées) :

1. Ibogaine & Noribogaine : Hybrid arrest, GDNF neuroplasticity, addiction reset
2. Resveratrol & SIRT1 : Minimal arrest (témoin négatif), échec seuil
3. Fasting & Breathing : Natural oscillators, physiological arrest principles
4. Psilocybin & LSD : High-entropy oscillation, DMN connectivity, TRD applications
5. AI Memory Extension : Dropout/consolidation parallels, computational arrest

Documentation améliorée :

• requirements.txt créé avec versions exactes (numpy 1.24.3, pandas 2.0.3, etc.)
• Références bibliographiques : 85 → 95+ sources
• Prédictions : 42 → 44 (ajout métabolites psychédéliques)

Statistiques mises à jour dans README :

• Compound_Properties_Database.csv : 10 lignes, 36 colonnes
• Literature sources : 95+ PMIDs
• Case studies : 5 (vs 0 précédemment)

---

v1.0.1 (2025-10-21) - Corrections Post-Rapport

Harmonisation prédictions (42→44) :

• Nombre de prédictions corrigé dans v6.txt pour correspondre au CSV
• Statistiques de confiance recalculées : High 18/44 (41%), Moderate 13/44 (30%), Low 13/44 (30%)

Fichiers manquants créés (5) :

• Experimental_Protocols_Summary.csv, R_Code_Figures_S2.R, Data_Dictionary.md
• Literature_Search_Strategy.md, API_Calculations_Full.xlsx

Figures brouillons créés (3) :

• Figure_S1_Molecular_Structures_draft.png, Figure_S2_Oscillatory_Advantage_draft.png/tiff
• Figure_S3_API_Flowchart_draft.png

Améliorations code :

• Option --random-seed ajoutée au script Python (défaut 42 reproductible)

---

v1.0 (2025-10-21) - Initial Release

2025-10-21: Dataset created, v1.0 submitted with manuscript