LabsHybrid45 credits

Lead Optimization Lab — AI Drug Analog Design

Turn a known bioactive compound into patentable analogs with improved drug properties

The Lead Optimization Lab takes a lead compound (SMILES) and generates hundreds of novel analogs optimized for potency, drug-likeness, safety, synthesizability, and patentability. It chains 8 computational steps — pharmacophore analysis, REINVENT4 generative chemistry, Tanimoto novelty scoring, DiffDock docking, ADMET profiling, and multi-objective ranking — into a single intelligent pipeline. Returns ranked candidates with SAR maps, patent family groupings, and draft Markush claims.

$0.45

per API call

credits per call

/v1/labs/leadopt/optimize

API endpoint

Features

8-step automated pipeline from lead compound to ranked candidates

Three generation approaches: scaffold decoration, scaffold hopping, fragment growing

Multi-objective scoring: potency, selectivity, drug-likeness, ADMET, novelty, synthesizability, BBB

Automatic pharmacophore extraction and constraint generation

Tanimoto novelty scoring against parent compound and known analogs

Patent family grouping with draft Markush claim generation

SAR map showing which modifications improve which properties

Optional target-based docking with DiffDock or Vina

Quick Start

LeadOpt Pipeline (REINVENT4 + DiffDock + ADMET-AI + RDKit) — Python Examplepython

import requests

API_KEY = "sk-sci-your-key-here"
url = "https://scirouter.ai/v1/labs/leadopt/optimize"

# Optimize SMER28 (autophagy enhancer)
response = requests.post(url, json={
    "lead_smiles": "C=CCNc1ncnc2ccc(Br)cc12",
    "lead_name": "SMER28",
    "target_pdb": "1S3S",
    "optimization_goals": ["potency", "selectivity", "admet", "novelty"],
    "generation_config": {
        "approaches": ["scaffold_decoration", "scaffold_hopping"],
        "n_molecules_per_approach": 200
    }
}, headers={"Authorization": f"Bearer {API_KEY}"})

result = response.json()["data"]["result"]
print(f"Generated: {result['generation_summary']['total_generated']}")
print(f"Final candidates: {len(result['candidates'])}")
for c in result["candidates"][:5]:
    print(f"  {c['name']}: score={c['composite_score']:.2f}, "
          f"patent={c['novelty_assessment']['patent_distance']}")