Skip to content

Latest commit

 

History

History
341 lines (258 loc) · 7.48 KB

QUICKSTART.md

File metadata and controls

341 lines (258 loc) · 7.48 KB

ZeroPain v4.0 - Quick Start Guide

Professional Molecular Therapeutics Platform

🚀 Installation

1. Clone Repository

git clone https://github.com/SWORDIntel/ZEROPAIN.git
cd ZEROPAIN

2. Create Virtual Environment

python3 -m venv zeropain_env
source zeropain_env/bin/activate  # On Windows: zeropain_env\Scripts\activate

3. Install Base Requirements

# Install core dependencies
pip install -r requirements/base.txt

# Install web interface
pip install -r requirements/web.txt

4. Install Molecular Modeling (Optional but Recommended)

# RDKit requires conda
conda install -c conda-forge rdkit openbabel

# Or use pip for some components
pip install -r requirements/molecular.txt

5. Install Intel AI Acceleration (Optional)

# For Intel Arc GPU / NPU
pip install -r requirements/intel.txt

# Download and install OpenVINO
# https://www.intel.com/content/www/us/en/developer/tools/openvino-toolkit/download.html

🎯 Quick Test

Test Molecular Analysis

python zeropain/molecular/structure.py

Test Intel AI Inference

python zeropain/molecular/intel_ai.py

Test Docking Module

python zeropain/molecular/docking.py

🌐 Launch Web Interface

1. Start API Backend

cd zeropain/api
python main.py

API will be available at:

2. Open Web Frontend

# Open in browser
open web/frontend/public/index.html

# Or use Python server
cd web/frontend/public
python -m http.server 3000
# Then visit: http://localhost:3000

📊 Example Usage

Example 1: Analyze a Compound

from zeropain.molecular.structure import from_smiles

# Morphine
morphine_smiles = "CN1CC[C@]23[C@H]4Oc5c(O)ccc(C[C@@H]1[C@@H]2C=C[C@@H]4O)c35"
struct = from_smiles(morphine_smiles, "Morphine", generate_3d=True)

print(f"Molecular Weight: {struct.mol_weight:.2f}")
print(f"LogP: {struct.logp:.2f}")
print(f"Drug-like: {struct.is_drug_like()}")

Example 2: Molecular Docking

from zeropain.molecular.docking import AutoDockVina

docking = AutoDockVina()
result = docking.dock(
    ligand_smiles=morphine_smiles,
    compound_name="Morphine",
    receptor="MOR"
)

print(f"Binding Affinity: {result.binding_affinity:.2f} kcal/mol")
print(f"Predicted Ki: {result.ki_predicted:.2f} nM")

Example 3: Intel AI ADMET Prediction

from zeropain.molecular.intel_ai import IntelAIMolecularPredictor
from zeropain.molecular.structure import from_smiles

predictor = IntelAIMolecularPredictor()

struct = from_smiles(morphine_smiles, "Morphine")
admet = predictor.predict_admet(struct.to_dict())

print(f"Bioavailability: {admet.bioavailability*100:.1f}%")
print(f"Half-life: {admet.half_life:.1f} hours")
print(f"BBB Permeability: {admet.bbb_permeability*100:.1f}%")

Example 4: API Usage

import requests

# Analyze molecule via API
response = requests.post(
    "http://localhost:8000/api/molecular/analyze",
    json={
        "name": "Morphine",
        "smiles": morphine_smiles
    }
)

data = response.json()
print(f"Molecular Weight: {data['molecular_weight']}")

🧪 Run Complete Pipeline

Using Python API

from zeropain.molecular.docking import AutoDockVina
from zeropain.molecular.intel_ai import IntelAIMolecularPredictor
from zeropain.molecular.structure import from_smiles

# 1. Analyze structure
struct = from_smiles(your_smiles, "MyCompound")

# 2. Predict ADMET
predictor = IntelAIMolecularPredictor()
admet = predictor.predict_admet(struct.to_dict())

# 3. Dock to receptor
docking = AutoDockVina()
result = docking.dock(your_smiles, "MyCompound", "MOR")

# 4. Analyze results
print(f"Safety Score: {struct.calculate_safety_score()}")
print(f"Binding Affinity: {result.binding_affinity} kcal/mol")
print(f"Bioavailability: {admet.bioavailability*100}%")

🎨 Web Interface Features

TEMPEST Class C Theme

  • Dark tactical interface with grid overlay
  • Real-time status indicators
  • High-contrast data visualization
  • Secure, classified aesthetic

Available Modules

  1. Compound Browser - Search and analyze compounds
  2. Molecular Docking - Protein-ligand binding prediction
  3. Protocol Optimization - Multi-compound protocol design
  4. Patient Simulation - Large-scale population studies
  5. Data Analysis - Statistical analysis and reporting
  6. Intel AI Inference - ADMET and toxicity prediction

🔧 Configuration

Intel Acceleration

Create .env file:

USE_INTEL_ACCELERATION=true
OPENVINO_DEVICE=AUTO  # AUTO, NPU, GPU, CPU

API Configuration

API_HOST=0.0.0.0
API_PORT=8000
WORKERS=4
LOG_LEVEL=info

📦 Download Receptor Structures

# Download PDB files for opioid receptors
cd data/receptors

# μ-opioid receptor (MOR) - PDB: 4DKL
wget https://files.rcsb.org/download/4DKL.pdb

# δ-opioid receptor (DOR) - PDB: 4N6H
wget https://files.rcsb.org/download/4N6H.pdb

# κ-opioid receptor (KOR) - PDB: 4DJH
wget https://files.rcsb.org/download/4DJH.pdb

# Convert to PDBQT format (requires MGLTools or Open Babel)
# obabel 4DKL.pdb -O mor.pdbqt -p 7.4

🧬 Add Custom Compounds

Via Python

from zeropain.core.compounds import CompoundProfile

custom = CompoundProfile(
    name="MyCompound",
    ki_orthosteric=25.0,  # nM
    ki_allosteric1=float('inf'),
    ki_allosteric2=float('inf'),
    g_protein_bias=6.0,
    beta_arrestin_bias=0.3,
    t_half=8.0,  # hours
    bioavailability=0.75,
    intrinsic_activity=0.45,
    tolerance_rate=0.2,
    prevents_withdrawal=True
)

# Save to database
custom.save()

Via Web Interface

  1. Navigate to "Compound Browser"
  2. Click "Add Custom Compound"
  3. Enter molecular properties
  4. Upload SMILES or draw structure
  5. Run docking and ADMET prediction
  6. Save to database

📊 Performance Benchmarks

On 16-core system with Intel NPU:

  • Molecular Docking: < 5 minutes per compound
  • ADMET Prediction: 1000+ predictions/second
  • Patient Simulation: 100k patients in ~2 minutes
  • Protocol Optimization: 1000 patients/second

🆘 Troubleshooting

RDKit Not Found

# Install via conda (recommended)
conda install -c conda-forge rdkit

# Or use pip (limited features)
pip install rdkit-pypi

AutoDock Vina Not Found

# Download from GitHub
wget https://github.com/ccsb-scripps/AutoDock-Vina/releases/download/v1.2.5/vina_1.2.5_linux_x86_64
chmod +x vina_1.2.5_linux_x86_64
sudo mv vina_1.2.5_linux_x86_64 /usr/local/bin/vina

Intel Optimizations Not Working

  • Verify Intel Extension for PyTorch: python -c "import intel_extension_for_pytorch as ipex; print(ipex.__version__)"
  • Check OpenVINO: python -c "from openvino.runtime import Core; print(Core().available_devices)"
  • System will fall back to CPU if hardware not detected

API Not Starting

# Check if port 8000 is available
lsof -i :8000

# Use different port
uvicorn zeropain.api.main:app --port 8080

📚 Next Steps

  1. Read Full Documentation: docs/api/
  2. Try Tutorial Notebooks: docs/tutorials/
  3. View Example Workflows: docs/guides/
  4. Join Development: See CONTRIBUTING.md

🔐 Security Note

This is a TEMPEST Class C system. Ensure:

  • No unauthorized access
  • Encrypted data transmission
  • Secure compound database
  • Audit logging enabled

ZeroPain Therapeutics v4.0 Zero Addiction • Zero Tolerance • Zero Withdrawal