File size: 8,707 Bytes
adecc9b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
"""
Phase 2 Verification Script
=============================

Tests the real open-source encoder implementations.

Usage:
    python scripts/verify_phase2.py [--full]
    
    --full: Run full tests with model downloads (slow, requires GPU recommended)
    
Without --full, runs quick tests with fallback/mock behavior.
"""

import sys
import os
from pathlib import Path

# Add project root
sys.path.insert(0, str(Path(__file__).parent.parent))

import logging
logging.basicConfig(level=logging.INFO, format='%(levelname)s: %(message)s')
logger = logging.getLogger(__name__)


def test_registry():
    """Test that all plugins are importable and registry works."""
    print("\n" + "="*60)
    print("🛠️  TEST 1: Plugin Registry")
    print("="*60)
    
    from bioflow.core import ToolRegistry, Modality
    from bioflow.plugins import (
        OBMEncoder,
        TextEncoder,
        MoleculeEncoder,
        ProteinEncoder,
        QdrantRetriever,
        DeepPurposePredictor
    )
    
    print("✅ All plugins imported successfully")
    
    # List available
    print("\nAvailable plugins:")
    print("  • OBMEncoder (multimodal)")
    print("  • TextEncoder (PubMedBERT/SciBERT)")
    print("  • MoleculeEncoder (ChemBERTa/RDKit)")
    print("  • ProteinEncoder (ESM-2/ProtBERT)")
    print("  • QdrantRetriever (Qdrant)")
    print("  • DeepPurposePredictor (DTI)")
    
    return True


def test_rdkit_fallback():
    """Test RDKit molecule encoding (no GPU needed)."""
    print("\n" + "="*60)
    print("🧪 TEST 2: RDKit Molecule Encoder (CPU-only)")
    print("="*60)
    
    try:
        from bioflow.plugins.encoders.molecule_encoder import MoleculeEncoder
        from bioflow.core import Modality
        
        encoder = MoleculeEncoder(backend="rdkit_morgan", fp_size=2048)
        
        test_molecules = [
            ("CCO", "Ethanol"),
            ("CC(=O)Oc1ccccc1C(=O)O", "Aspirin"),
            ("CN1C=NC2=C1C(=O)N(C(=O)N2C)C", "Caffeine"),
        ]
        
        print(f"Encoder dimension: {encoder.dimension}")
        print("\nEncoding molecules:")
        
        for smiles, name in test_molecules:
            result = encoder.encode(smiles, Modality.SMILES)
            nonzero = sum(1 for v in result.vector if v > 0)
            print(f"  • {name}: {nonzero} bits set (of {len(result.vector)})")
        
        print("✅ RDKit encoding works!")
        return True
        
    except ImportError as e:
        print(f"⚠️  RDKit not installed: {e}")
        print("   Install with: pip install rdkit")
        return False


def test_deeppurpose_predictor():
    """Test DeepPurpose predictor (with fallback)."""
    print("\n" + "="*60)
    print("🔮 TEST 3: DeepPurpose Predictor")
    print("="*60)
    
    from bioflow.plugins.deeppurpose_predictor import DeepPurposePredictor
    
    predictor = DeepPurposePredictor()
    
    # Test data
    drug = "CC(=O)Oc1ccccc1C(=O)O"  # Aspirin
    target = "MKTVRQERLKSIVRILERSKEPVSGAQLAEELSVSRQVIVQDIAYLRSLGYNIVATPRGYVLAGG"
    
    print(f"Drug: Aspirin (SMILES: {drug[:30]}...)")
    print(f"Target: Protein ({len(target)} amino acids)")
    
    result = predictor.predict(drug, target)
    
    print(f"\nPrediction:")
    print(f"  • Score: {result.score:.3f}")
    print(f"  • Label: {result.label}")
    print(f"  • Confidence: {result.confidence:.2f}")
    print(f"  • Method: {result.metadata.get('method', 'unknown')}")
    
    if result.metadata.get('warning'):
        print(f"  ⚠️  {result.metadata['warning']}")
    
    print("✅ Predictor works (with fallback if DeepPurpose unavailable)")
    return True


def test_qdrant_retriever():
    """Test Qdrant retriever with mock encoder."""
    print("\n" + "="*60)
    print("🗄️  TEST 4: Qdrant Retriever (In-Memory)")
    print("="*60)
    
    try:
        from bioflow.core import BioEncoder, Modality, EmbeddingResult
        from bioflow.plugins.qdrant_retriever import QdrantRetriever
        
        # Mock encoder for testing
        class MockEncoder(BioEncoder):
            def encode(self, content, modality):
                # Simple hash-based vector
                import hashlib
                h = hashlib.md5(content.encode()).hexdigest()
                vector = [int(c, 16) / 15.0 for c in h] * 48  # 768-dim
                return EmbeddingResult(vector=vector[:768], modality=modality, dimension=768)
            
            @property
            def dimension(self): return 768
        
        encoder = MockEncoder()
        retriever = QdrantRetriever(encoder=encoder, collection="test_molecules")
        
        # Ingest test data
        test_data = [
            ("CCO", "Ethanol", {"type": "alcohol"}),
            ("CCCO", "Propanol", {"type": "alcohol"}),
            ("CC(=O)O", "Acetic acid", {"type": "acid"}),
            ("c1ccccc1", "Benzene", {"type": "aromatic"}),
        ]
        
        print("Ingesting molecules...")
        for smiles, name, payload in test_data:
            retriever.ingest(smiles, Modality.SMILES, {"name": name, **payload})
        
        print(f"Collection size: {retriever.count()}")
        
        # Search
        print("\nSearching for 'CCCCO' (Butanol)...")
        results = retriever.search("CCCCO", limit=3, modality=Modality.SMILES)
        
        print("Results:")
        for r in results:
            print(f"  • {r.payload.get('name', 'Unknown')}: score={r.score:.3f}")
        
        print("✅ Qdrant retriever works!")
        return True
        
    except ImportError as e:
        print(f"⚠️  qdrant-client not installed: {e}")
        print("   Install with: pip install qdrant-client")
        return False


def test_full_obm_encoder():
    """Test full OBM encoder with real models (slow, requires downloads)."""
    print("\n" + "="*60)
    print("🚀 TEST 5: Full OBM Encoder (requires model downloads)")
    print("="*60)
    
    try:
        from bioflow.plugins.obm_encoder import OBMEncoder
        from bioflow.core import Modality
        
        print("Initializing OBMEncoder...")
        print("(This will download models on first run - ~500MB)")
        
        obm = OBMEncoder(
            text_model="pubmedbert",
            molecule_model="chemberta",
            protein_model="esm2_t6",  # Smallest ESM model
            lazy_load=True
        )
        
        # Test text
        print("\n1. Encoding text...")
        text_result = obm.encode("EGFR inhibitor for lung cancer treatment", Modality.TEXT)
        print(f"   Text embedding: {len(text_result.vector)} dims")
        
        # Test molecule
        print("2. Encoding molecule...")
        mol_result = obm.encode("CC(=O)Oc1ccccc1C(=O)O", Modality.SMILES)
        print(f"   Molecule embedding: {len(mol_result.vector)} dims")
        
        # Test protein
        print("3. Encoding protein...")
        prot_result = obm.encode("MKTVRQERLKSIVRILERSKEPVSG", Modality.PROTEIN)
        print(f"   Protein embedding: {len(prot_result.vector)} dims")
        
        # Cross-modal similarity
        print("\n4. Cross-modal similarity:")
        sim = obm.similarity(text_result, mol_result)
        print(f"   Text-Molecule similarity: {sim:.3f}")
        
        print("\n✅ Full OBM Encoder works!")
        print(obm.get_encoder_info())
        return True
        
    except Exception as e:
        print(f"❌ Error: {e}")
        import traceback
        traceback.print_exc()
        return False


def main():
    """Run verification tests."""
    print("="*60)
    print("🧬 BioFlow Phase 2 Verification")
    print("="*60)
    
    full_mode = "--full" in sys.argv
    
    if full_mode:
        print("Running FULL tests (with model downloads)")
    else:
        print("Running QUICK tests (no model downloads)")
        print("Add --full flag for complete testing")
    
    results = {}
    
    # Always run
    results["Registry"] = test_registry()
    results["RDKit"] = test_rdkit_fallback()
    results["DeepPurpose"] = test_deeppurpose_predictor()
    results["Qdrant"] = test_qdrant_retriever()
    
    # Only in full mode
    if full_mode:
        results["OBMEncoder"] = test_full_obm_encoder()
    
    # Summary
    print("\n" + "="*60)
    print("📊 VERIFICATION SUMMARY")
    print("="*60)
    
    for test, passed in results.items():
        status = "✅ PASS" if passed else "❌ FAIL"
        print(f"  {test}: {status}")
    
    all_passed = all(results.values())
    print("\n" + ("✅ All tests passed!" if all_passed else "⚠️  Some tests failed"))
    
    return 0 if all_passed else 1


if __name__ == "__main__":
    sys.exit(main())