undefined

---

Phase 0: Tool Verification (check parameter names)
    ↓
Phase 1: Target Validation
    ├─ 1.1 Resolve identifiers (UniProt, Ensembl, ChEMBL target ID)
    ├─ 1.2 Assess druggability/tractability
    │   └─ 1.2.5 Check therapeutic antibodies (Thera-SAbDab) [NEW]
    ├─ 1.3 Identify binding sites
    └─ 1.4 Predict structure (NvidiaNIM_alphafold2/esmfold)
    ↓
Phase 2: Known Ligand Mining
    ├─ Extract ChEMBL bioactivity data
    ├─ Get GtoPdb interactions
    ├─ Identify chemical probes
    ├─ BindingDB affinity data (NEW - Ki/IC50/Kd)
    ├─ PubChem BioAssay HTS data (NEW - screening hits)
    └─ Analyze SAR from known actives
    ↓
Phase 3: Structure Analysis
    ├─ Get PDB structures with ligands
    ├─ Check EMDB for cryo-EM structures (NEW - for membrane targets)
    ├─ Analyze binding pocket
    └─ Identify key interactions
    ↓
Phase 3.5: Docking Validation (NvidiaNIM_diffdock/boltz2) [NEW]
    ├─ Dock reference inhibitor
    └─ Validate binding pocket geometry
    ↓
Phase 4: Compound Expansion
    ├─ 4.1-4.3 Similarity/substructure search
    └─ 4.4 De novo generation (NvidiaNIM_genmol/molmim) [NEW]
    ↓
Phase 5: ADMET Filtering
    ├─ Predict physicochemical properties
    ├─ Predict ADMET endpoints
    └─ Flag liabilities
    ↓
Phase 6: Candidate Docking & Prioritization
    ├─ Dock all candidates (NvidiaNIM_diffdock/boltz2) [UPDATED]
    ├─ Score by docking + ADMET + novelty
    ├─ Assess synthesis feasibility
    └─ Generate final ranked list
    ↓
Phase 7: Report Synthesis

undefined

undefined

1. UniProt_search(query=target_name, organism="human")
   └─ Extract: UniProt accession, gene name, protein name

2. MyGene_query_genes(q=gene_symbol, species="human")
   └─ Extract: Ensembl gene ID, NCBI gene ID

3. ChEMBL_search_targets(query=target_name, organism="Homo sapiens")
   └─ Extract: ChEMBL target ID, target type

4. GtoPdb_get_targets(query=target_name)
   └─ Extract: GtoPdb target ID (if GPCR/ion channel/enzyme)

ids = {
    'uniprot': 'P00533',
    'ensembl': 'ENSG00000146648',
    'chembl_target': 'CHEMBL203',
    'gene_symbol': 'EGFR',
    'gtopdb': '1797'  # if available
}

1. OpenTargets_get_target_tractability_by_ensemblID(ensemblId)
   └─ Extract: Small molecule tractability score, bucket
   
2. DGIdb_get_gene_druggability(genes=[gene_symbol])
   └─ Extract: Druggability categories, known drug count
   
3. OpenTargets_get_target_classes_by_ensemblID(ensemblId)
   └─ Extract: Target class (kinase, GPCR, etc.)

4. GPCRdb_get_protein(protein=entry_name)  # NEW - for GPCRs
   └─ Extract: GPCR family, receptor state, ligand binding data

def check_if_gpcr_and_enrich(tu, target_name, uniprot_id):
    """Check if target is GPCR and get specialized data."""
    
    # Build GPCRdb entry name (e.g., "adrb2_human")
    entry_name = f"{target_name.lower()}_human"
    
    # Check if it's a GPCR
    gpcr_info = tu.tools.GPCRdb_get_protein(
        operation="get_protein",
        protein=entry_name
    )
    
    if gpcr_info.get('status') == 'success':
        # It's a GPCR - get specialized data
        
        # Get known structures (active/inactive states)
        structures = tu.tools.GPCRdb_get_structures(
            operation="get_structures",
            protein=entry_name
        )
        
        # Get known ligands
        ligands = tu.tools.GPCRdb_get_ligands(
            operation="get_ligands",
            protein=entry_name
        )
        
        # Get mutation data (important for SAR)
        mutations = tu.tools.GPCRdb_get_mutations(
            operation="get_mutations",
            protein=entry_name
        )
        
        return {
            'is_gpcr': True,
            'gpcr_family': gpcr_info['data'].get('family'),
            'gpcr_class': gpcr_info['data'].get('receptor_class'),
            'structures': structures['data'].get('structures', []),
            'ligands': ligands['data'].get('ligands', []),
            'mutation_data': mutations['data'].get('mutations', [])
        }
    
    return {'is_gpcr': False}

def check_therapeutic_antibodies(tu, target_name):
    """
    Check Thera-SAbDab for therapeutic antibodies against target.
    """
    # Search by target name
    results = tu.tools.TheraSAbDab_search_by_target(
        target=target_name
    )
    
    if results.get('status') == 'success':
        antibodies = results['data'].get('therapeutics', [])
        
        # Categorize by clinical stage
        by_phase = {'Approved': [], 'Phase 3': [], 'Phase 2': [], 'Phase 1': [], 'Preclinical': []}
        for ab in antibodies:
            phase = ab.get('phase', 'Unknown')
            for key in by_phase.keys():
                if key.lower() in phase.lower():
                    by_phase[key].append(ab)
                    break
        
        return {
            'total_antibodies': len(antibodies),
            'by_phase': by_phase,
            'antibodies': antibodies[:10],  # Top 10
            'competitive_alert': len(by_phase.get('Approved', [])) > 0
        }
    return None

def get_antibody_landscape(tu, target_name, uniprot_id=None):
    """
    Comprehensive antibody competitive landscape.
    """
    # Thera-SAbDab search
    therasabdab = check_therapeutic_antibodies(tu, target_name)
    
    # Also search by common synonyms
    synonyms = [target_name]
    if target_name != uniprot_id:
        synonyms.append(uniprot_id)
    
    all_antibodies = []
    for synonym in synonyms:
        results = tu.tools.TheraSAbDab_search_therapeutics(query=synonym)
        if results.get('status') == 'success':
            all_antibodies.extend(results['data'].get('therapeutics', []))
    
    # Deduplicate
    seen = set()
    unique = []
    for ab in all_antibodies:
        inn = ab.get('inn_name')
        if inn and inn not in seen:
            seen.add(inn)
            unique.append(ab)
    
    return {
        'antibodies': unique,
        'count': len(unique),
        'has_approved': any(ab.get('phase', '').lower() == 'approved' for ab in unique),
        'source': 'Thera-SAbDab'
    }

undefined

**Why Include Antibody Landscape**:
- **Validation**: Approved antibodies = validated target
- **Competition**: Understand what's already in market/clinic
- **Strategy**: Identify gaps (no oral, no CNS-penetrant)
- **Synergy**: Potential combination opportunities

1. ChEMBL_search_binding_sites(target_chembl_id)
   └─ Extract: Binding site names, types
   
2. get_binding_affinity_by_pdb_id(pdb_id)  # For each PDB with ligand
   └─ Extract: Kd, Ki, IC50 values for co-crystallized ligands
   
3. InterPro_get_protein_domains(uniprot_accession)
   └─ Extract: Domain architecture, active sites

undefined

undefined

NvidiaNIM_alphafold2(
    sequence=kinase_domain_sequence,
    algorithm="mmseqs2",
    relax_prediction=False
)
└─ Returns: PDB structure with pLDDT confidence scores
└─ Use when: Accuracy is critical, time is available (~5-15 min)

NvidiaNIM_esmfold(sequence=kinase_domain_sequence)
└─ Returns: PDB structure (max 1024 AA)
└─ Use when: Quick assessment needed (~30 sec)

undefined

---

1. ChEMBL_get_target_activities(target_chembl_id, limit=500)
   └─ Filter: standard_type in ["IC50", "Ki", "Kd", "EC50"]
   └─ Filter: standard_value < 10000 nM
   └─ Extract: ChEMBL molecule IDs, SMILES, potency values

2. ChEMBL_get_molecule(molecule_chembl_id)  # For top actives
   └─ Extract: Full molecular data, max_phase, oral flag

undefined

undefined

GtoPdb_get_target_interactions(target_id)
└─ Extract: Ligands with pKi/pIC50, selectivity data

OpenTargets_get_chemical_probes_by_target_ensemblID(ensemblId)
└─ Extract: Validated chemical probes with ratings

undefined

undefined

undefined

undefined

def get_bindingdb_ligands(tu, uniprot_id, affinity_cutoff=10000):
    """
    Get ligands from BindingDB with measured affinities.
    
    BindingDB advantages:
    - May have compounds not in ChEMBL
    - Different affinity types (Ki, IC50, Kd)
    - Direct literature links
    """
    
    result = tu.tools.BindingDB_get_ligands_by_uniprot(
        uniprot=uniprot_id,
        affinity_cutoff=affinity_cutoff  # nM
    )
    
    if result:
        ligands = []
        for entry in result:
            ligands.append({
                'smiles': entry.get('smile'),
                'affinity_type': entry.get('affinity_type'),
                'affinity_nM': entry.get('affinity'),
                'pmid': entry.get('pmid'),
                'monomer_id': entry.get('monomerid')
            })
        
        # Sort by potency
        ligands.sort(key=lambda x: float(x['affinity_nM']) if x['affinity_nM'] else 1e6)
        return ligands[:50]  # Top 50
    
    return []

def find_compound_polypharmacology(tu, smiles, similarity_cutoff=0.85):
    """Find off-target interactions for selectivity analysis."""
    
    targets = tu.tools.BindingDB_get_targets_by_compound(
        smiles=smiles,
        similarity_cutoff=similarity_cutoff
    )
    
    return targets  # Other proteins this compound may bind

undefined

undefined

def get_pubchem_assays_for_target(tu, gene_symbol):
    """
    Get bioassays and active compounds from PubChem.
    
    Advantages:
    - HTS data not in ChEMBL
    - NIH-funded screening programs (MLPCN)
    - Dose-response curves for IC50 calculation
    """
    
    # Search assays targeting this gene
    assays = tu.tools.PubChem_search_assays_by_target_gene(
        gene_symbol=gene_symbol
    )
    
    results = {
        'assays': [],
        'total_active_compounds': 0
    }
    
    if assays.get('data', {}).get('aids'):
        for aid in assays['data']['aids'][:10]:  # Top 10 assays
            # Get assay summary
            summary = tu.tools.PubChem_get_assay_summary(aid=aid)
            
            # Get active compounds
            actives = tu.tools.PubChem_get_assay_active_compounds(aid=aid)
            active_cids = actives.get('data', {}).get('cids', [])
            
            results['assays'].append({
                'aid': aid,
                'summary': summary.get('data', {}),
                'active_count': len(active_cids)
            })
            results['total_active_compounds'] += len(active_cids)
    
    return results

def get_dose_response_data(tu, aid):
    """Get dose-response curves for IC50/EC50 determination."""
    
    dr_data = tu.tools.PubChem_get_assay_dose_response(aid=aid)
    return dr_data

def get_compound_bioactivity_profile(tu, cid):
    """Get all bioactivity data for a compound."""
    
    profile = tu.tools.PubChem_get_compound_bioactivity(cid=cid)
    return profile

undefined

**Why Use Both BindingDB and PubChem**:
| Source | Strengths | Best For |
|--------|-----------|----------|
| **ChEMBL** | Curated, standardized, SAR data | Primary ligand source |
| **BindingDB** | Direct affinity measurements | Ki/Kd values, PMIDs |
| **PubChem BioAssay** | HTS data, NIH screens | Novel scaffolds, broad coverage |

---

1. PDB_search_similar_structures(query=uniprot_accession, type="sequence")
   └─ Extract: PDB IDs with ligands
   
2. get_protein_metadata_by_pdb_id(pdb_id)
   └─ Extract: Resolution, method, ligand codes
   
3. alphafold_get_prediction(accession=uniprot_accession)
   └─ Extract: Predicted structure (if no experimental)

def get_cryoem_structures(tu, target_name, uniprot_accession):
    """Get cryo-EM structures for membrane targets."""
    
    # Search EMDB
    emdb_results = tu.tools.emdb_search(
        query=f"{target_name} membrane receptor"
    )
    
    structures = []
    for entry in emdb_results[:5]:
        details = tu.tools.emdb_get_entry(entry_id=entry['emdb_id'])
        
        # Get associated PDB model (essential for docking)
        pdb_models = details.get('pdb_ids', [])
        
        structures.append({
            'emdb_id': entry['emdb_id'],
            'resolution': entry.get('resolution', 'N/A'),
            'title': entry.get('title', 'N/A'),
            'conformational_state': details.get('state', 'Unknown'),
            'pdb_models': pdb_models
        })
    
    return structures

undefined

undefined

get_binding_affinity_by_pdb_id(pdb_id)
└─ Extract: Binding affinities for co-crystallized ligands

undefined

---

NvidiaNIM_diffdock(
    protein=pdb_content,        # PDB text content
    ligand=reference_sdf,       # SDF/MOL2 content
    num_poses=10
)
└─ Returns: Docking poses with confidence scores
└─ Use: When you have PDB structure and ligand SDF file

NvidiaNIM_boltz2(
    polymers=[{"molecule_type": "protein", "sequence": kinase_sequence}],
    ligands=[{"smiles": "COc1cc2ncnc(Nc3ccc(C#C)cc3)c2cc1OCCOC"}],
    sampling_steps=50,
    diffusion_samples=1
)
└─ Returns: Protein-ligand complex structure
└─ Use: When starting from SMILES, no SDF needed

undefined

---

1. ChEMBL_search_similar_molecules(molecule=top_active_smiles, similarity=70)
   └─ Extract: Similar compounds not yet tested on target
   
2. PubChem_search_compounds_by_similarity(smiles, threshold=0.7)
   └─ Extract: PubChem CIDs with similar structures

1. ChEMBL_search_substructure(smiles=core_scaffold)
   └─ Extract: Compounds containing scaffold
   
2. PubChem_search_compounds_by_substructure(smiles=core_scaffold)
   └─ Extract: Additional scaffold-containing compounds

1. STITCH_get_chemical_protein_interactions(identifier=target_gene)
   └─ Extract: Additional chemical-protein links
   
2. DGIdb_get_drug_gene_interactions(genes=[gene_symbol])
   └─ Extract: Approved/investigational drugs

undefined

undefined

NvidiaNIM_genmol(
    smiles="COc1cc2ncnc(Nc3ccc([*{3-8}])c([*{1-3}])c3)c2cc1OCCCN1CCOCC1",
    num_molecules=100,
    temperature=2.0,
    scoring="QED"
)
└─ Input: SMILES with [*{min-max}] masked regions
└─ Output: Generated molecules with QED/LogP scores
└─ Use: Explore specific positions while keeping scaffold

undefined

**Option B: MolMIM (Controlled Generation from Reference)**

**Generation Workflow**:
1. Identify top 3-5 actives from Phase 2
2. Design masked SMILES for GenMol OR use as reference for MolMIM
3. Generate 50-100 molecules per seed
4. Pass generated molecules to Phase 5 (ADMET filtering)
5. Dock survivors in Phase 6 for final ranking

**Report Format**:
```markdown

---

ADMETAI_predict_physicochemical_properties(smiles=[compound_list])
└─ Filter: Lipinski violations ≤ 1
└─ Filter: QED > 0.3
└─ Filter: MW 200-600

1. ADMETAI_predict_bioavailability(smiles=[compound_list])
   └─ Filter: Oral bioavailability > 0.3
   
2. ADMETAI_predict_toxicity(smiles=[compound_list])
   └─ Filter: AMES < 0.5, hERG < 0.5, DILI < 0.5
   
3. ADMETAI_predict_CYP_interactions(smiles=[compound_list])
   └─ Flag: CYP3A4 inhibitors (drug interaction risk)

ChEMBL_search_compound_structural_alerts(smiles=compound_smiles)
└─ Flag: PAINS, reactive groups, toxicophores

undefined

---

undefined

undefined

undefined

---

def search_binder_literature(tu, target_name, compound_scaffolds):
    """Search literature for compound and target evidence."""
    
    # PubMed: Published SAR studies
    sar_papers = tu.tools.PubMed_search_articles(
        query=f"{target_name} inhibitor SAR structure-activity",
        limit=30
    )
    
    # BioRxiv: Latest unpublished findings
    preprints = tu.tools.BioRxiv_search_preprints(
        query=f"{target_name} small molecule discovery",
        limit=15
    )
    
    # MedRxiv: Clinical data on inhibitors
    clinical = tu.tools.MedRxiv_search_preprints(
        query=f"{target_name} inhibitor clinical trial",
        limit=10
    )
    
    # Citation analysis for key papers
    key_papers = sar_papers[:10]
    for paper in key_papers:
        citation = tu.tools.openalex_search_works(
            query=paper['title'],
            limit=1
        )
        paper['citations'] = citation[0].get('cited_by_count', 0) if citation else 0
    
    return {
        'published_sar': sar_papers,
        'preprints': preprints,
        'clinical_preprints': clinical,
        'high_impact_papers': sorted(key_papers, key=lambda x: x.get('citations', 0), reverse=True)
    }

undefined

---

undefined

---