File size: 15,146 Bytes
4a8b134 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 | """
FastAPI Application for Drug Repurposing System
"""
import logging
from fastapi import FastAPI, HTTPException, Request
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import JSONResponse
import time
from app.config import settings
from app.models import (
DiseaseSearchRequest,
ScreeningRequest,
ScreeningResponse,
DrugCandidate,
HealthCheckResponse,
ErrorResponse,
EnrichedTargetResponse
)
from app.pipelines.screening_cache import load_cached_screening, save_cached_screening
from app.pipelines import (
DiseaseTargetPipeline,
ProteinSequencePipeline,
DrugLibraryPipeline,
AIScreeningPipeline,
ResultProcessingPipeline,
PdbStructurePipeline
)
# Configure logging
logging.basicConfig(
level=settings.LOG_LEVEL,
format=settings.LOG_FORMAT
)
logger = logging.getLogger(__name__)
# Initialize FastAPI app
app = FastAPI(
title=settings.API_TITLE,
description=settings.APP_DESCRIPTION,
version=settings.API_VERSION,
docs_url="/docs",
openapi_url="/openapi.json"
)
# Add CORS middleware
app.add_middleware(
CORSMiddleware,
allow_origins=settings.CORS_ORIGINS,
allow_credentials=settings.CORS_CREDENTIALS,
allow_methods=settings.CORS_METHODS,
allow_headers=settings.CORS_HEADERS,
)
# Initialize pipelines
disease_pipeline = DiseaseTargetPipeline(api_url=settings.OPENTARGETS_API_URL)
protein_pipeline = ProteinSequencePipeline(api_url=settings.UNIPROT_API_URL)
drug_pipeline = DrugLibraryPipeline(use_mock=settings.USE_MOCK_DRUGS)
ai_pipeline = AIScreeningPipeline(use_mock=settings.USE_MOCK_MODEL)
result_pipeline = ResultProcessingPipeline()
pdb_pipeline = PdbStructurePipeline(api_url=settings.UNIPROT_API_URL)
# Load AI model at startup
@app.on_event("startup")
async def startup_event():
"""Initialize AI model and drug library at startup - PRODUCTION MODE"""
logger.info(f"{'='*70}")
logger.info(f"Starting {settings.APP_NAME} v{settings.APP_VERSION}")
logger.info(f"{'='*70}")
logger.info(f"Device: {settings.DEVICE} {'(GPU)' if settings.HAS_GPU else '(CPU)'}")
logger.info(f"Max drugs per screening: {settings.MAX_DRUGS_FOR_DEMO}")
logger.info(f"Model: {settings.DEEP_PURPOSE_MODEL}")
logger.info(f"Dataset: {settings.TDC_DATASET}")
try:
if ai_pipeline.use_mock:
logger.info("\n[1/2] Mock mode enabled - skipping DeepPurpose model loading")
logger.info("[2/2] Mock drug library enabled - skipping TDC download")
logger.info(f"\n{'='*70}")
logger.info("β
MOCK MODE: All systems ready")
logger.info(" - Using mock predictions (random scores)")
logger.info(" - Drug library: local fallback")
logger.info(f"{'='*70}\n")
else:
logger.info("\n[1/2] Loading DeepPurpose MPNN_CNN_BindingDB model...")
ai_pipeline.load_model(model_name=settings.DEEP_PURPOSE_MODEL)
logger.info("β
AI model loaded successfully\n")
logger.info("[2/2] Verifying drug library...")
test_drugs = drug_pipeline.load_drug_library(
dataset_name=settings.TDC_DATASET,
max_drugs=10
)
logger.info(f"β
Drug library ready - {len(test_drugs)} sample drugs loaded\n")
logger.info(f"{'='*70}")
logger.info("β
PRODUCTION MODE: All systems ready")
logger.info(" - Real DeepPurpose MPNN_CNN predictions enabled")
logger.info(" - Drug library enabled (Official TDC or Local Fallback)")
logger.info(" - No mock predictions active")
logger.info(f"{'='*70}\n")
except ImportError as e:
logger.error(f"\n{'='*70}")
logger.error("β STARTUP FAILED: Missing required dependencies")
logger.error(f"{'='*70}")
logger.error(f"Error: {str(e)}\n")
if not ai_pipeline.use_mock:
logger.error("Install required packages:\n")
logger.error(" DeepPurpose and dependencies are REQUIRED")
logger.error(" Check terminal output above for pip install commands\n")
raise
except RuntimeError as e:
logger.error(f"\n{'='*70}")
logger.error("β STARTUP FAILED: Runtime error")
logger.error(f"{'='*70}")
logger.error(f"Error: {str(e)}\n")
raise
except Exception as e:
logger.error(f"\n{'='*70}")
logger.error(f"β STARTUP FAILED: Unexpected error")
logger.error(f"{'='*70}")
logger.error(f"Error: {str(e)}\n")
if not ai_pipeline.use_mock:
import traceback
traceback.print_exc()
raise
# Health Check Endpoint
@app.get("/health", response_model=HealthCheckResponse)
async def health_check():
"""Check API health status"""
return HealthCheckResponse(
status="healthy",
version=settings.APP_VERSION,
service=settings.APP_NAME
)
# Model Status Endpoint
@app.get("/api/v1/model-status")
async def model_status():
"""Get information about the loaded AI model"""
model_info = ai_pipeline.get_model_info()
return {
"model": model_info['model_name'],
"device": model_info['device'],
"gpu_available": model_info['gpu_available'],
"model_loaded": model_info['is_loaded'],
"using_mock_mode": model_info['using_mock'],
"batch_size": model_info['batch_size'],
"max_drugs_per_screening": settings.MAX_DRUGS_FOR_DEMO,
"version": settings.APP_VERSION
}
# Root Endpoint
@app.get("/")
async def root():
"""Root endpoint with API information"""
return {
"name": settings.APP_NAME,
"version": settings.APP_VERSION,
"description": settings.APP_DESCRIPTION,
"docs": "/docs",
"health": "/health"
}
# Disease Targets Endpoint (enriched with sequences + PDB IDs)
@app.post("/api/v1/disease-targets", response_model=EnrichedTargetResponse)
async def get_disease_targets(request: DiseaseSearchRequest):
"""
Get target proteins associated with a disease using Open Targets API,
enriched with protein sequences (UniProt) and PDB structure IDs.
- **disease_name**: Name of the disease to search for
- **top_n**: Number of top targets to retrieve (1-100, default: 10)
"""
try:
logger.info(f"Searching targets for disease: {request.disease_name}")
disease_id = disease_pipeline.fetch_disease_id(request.disease_name)
targets = disease_pipeline.get_disease_targets(
disease_name=request.disease_name,
top_n=request.top_n
)
if not targets:
raise HTTPException(
status_code=404,
detail=f"No targets found for disease: {request.disease_name}"
)
targets_with_seqs = protein_pipeline.get_protein_sequences(targets)
if not targets_with_seqs:
raise HTTPException(
status_code=400,
detail="Could not fetch sequences for any targets"
)
enriched = pdb_pipeline.fetch_pdb_ids(targets_with_seqs)
return {
"disease": request.disease_name,
"disease_id": disease_id or "unknown",
"total_targets": len(enriched),
"targets": enriched
}
except HTTPException:
raise
except Exception as e:
logger.error(f"Error getting disease targets: {str(e)}")
raise HTTPException(status_code=500, detail=str(e))
# Protein Sequences Endpoint
@app.post("/api/v1/protein-sequences")
async def get_protein_sequences(targets: list):
"""
Fetch protein sequences from UniProt for given target symbols.
- **targets**: List of target objects with 'symbol' field
"""
try:
logger.info(f"Fetching sequences for {len(targets)} targets")
sequences = protein_pipeline.get_protein_sequences(targets)
return {
"total_requested": len(targets),
"total_found": len(sequences),
"targets": sequences
}
except Exception as e:
logger.error(f"Error getting protein sequences: {str(e)}")
raise HTTPException(status_code=500, detail=str(e))
# Drug Library Endpoint
@app.get("/api/v1/drug-library")
async def get_drug_library():
"""
Load FDA-approved drug library with SMILES strings.
"""
try:
logger.info("Loading drug library")
drugs = drug_pipeline.load_drug_library(dataset_name=settings.TDC_DATASET)
# Limit number of drugs for demo
if len(drugs) > settings.MAX_DRUGS_FOR_DEMO:
drugs = drugs[:settings.MAX_DRUGS_FOR_DEMO]
logger.info(f"Limited drugs to {settings.MAX_DRUGS_FOR_DEMO} for demo")
return {
"total_drugs": len(drugs),
"drugs": drugs
}
except Exception as e:
logger.error(f"Error loading drug library: {str(e)}")
raise HTTPException(status_code=500, detail=str(e))
# Virtual Screening Endpoint (Main Pipeline)
@app.post("/api/v1/screen", response_model=ScreeningResponse)
async def virtual_screening(request: ScreeningRequest):
"""
Run complete virtual drug screening pipeline for a disease.
This endpoint runs the entire A-to-Z pipeline:
1. Identify disease targets (Open Targets)
2. Fetch protein sequences (UniProt)
3. Load drug library (TDC)
4. Run AI predictions (DeepPurpose)
5. Process and rank results
"""
try:
# Check cache
cached = load_cached_screening(request)
if cached is not None:
logger.info(f"Returning cached screening for: {request.disease_name}")
return ScreeningResponse(**cached)
start_time = time.time()
logger.info(f"Starting screening for disease: {request.disease_name}")
# Stage 1: Get disease targets
logger.info("Stage 1: Identifying disease targets...")
targets = disease_pipeline.get_disease_targets(
disease_name=request.disease_name,
top_n=min(request.top_n_targets, settings.MAX_TARGETS)
)
if not targets:
raise HTTPException(
status_code=404,
detail=f"No targets found for disease: {request.disease_name}"
)
# Stage 2: Get protein sequences
logger.info(f"Stage 2: Fetching sequences for {len(targets)} targets...")
targets_with_seqs = protein_pipeline.get_protein_sequences(targets)
if not targets_with_seqs:
raise HTTPException(
status_code=400,
detail="Could not fetch sequences for any targets"
)
# Stage 3: Load drug library
logger.info(f"Stage 3: Loading drug library (max {settings.MAX_DRUGS_FOR_DEMO} drugs)...")
drugs = drug_pipeline.load_drug_library(
dataset_name=settings.TDC_DATASET,
max_drugs=settings.MAX_DRUGS_FOR_DEMO
)
if not drugs:
raise HTTPException(
status_code=400,
detail="Could not load any drugs from library"
)
# Limit drugs based on GPU availability (performance optimization)
actual_max = settings.MAX_DRUGS_FOR_DEMO
if len(drugs) > actual_max:
logger.info(f"Limiting drugs to {actual_max} for performance")
drugs = drugs[:actual_max]
logger.info(f"Using {len(drugs)} drugs for screening")
# Stage 4: Run AI screening
n_pairs = len(drugs) * len(targets_with_seqs)
logger.info(f"Stage 4: Running AI predictions...")
logger.info(f" Drug-target pairs: {n_pairs} ({len(drugs)} drugs Γ {len(targets_with_seqs)} targets)")
if settings.HAS_GPU:
logger.info(f" GPU acceleration enabled (batch size: {settings.BATCH_SIZE})")
else:
logger.info(f" CPU mode (batch size: {settings.BATCH_SIZE})")
candidates, warnings = ai_pipeline.run_virtual_screening(drugs, targets_with_seqs)
if not candidates:
raise HTTPException(
status_code=400,
detail="AI screening failed to produce predictions"
)
# Stage 5: Process results
logger.info("Stage 5: Processing and ranking results...")
final_results = result_pipeline.process_final_results(
candidates,
known_drugs=request.known_drugs,
min_score=request.min_score
)
top_k = final_results[:10]
top_candidates = [
DrugCandidate(
drug_name=r["drug_name"],
smiles=r.get("smiles", ""),
target_symbol=r["target_symbol"],
uniprot_id=r.get("uniprot_id", ""),
binding_score=r["score"],
rank=i + 1,
status=r.get("status", "").replace("β
", "").replace("π ", "")
)
for i, r in enumerate(top_k)
]
elapsed_time = time.time() - start_time
logger.info(f"β
Screening complete in {elapsed_time:.2f} seconds")
logger.info(f" Total candidates: {len(final_results)}")
logger.info(f" Top candidates: {len(top_k)}")
response = ScreeningResponse(
disease_name=request.disease_name,
total_targets_found=len(targets_with_seqs),
total_drugs_screened=len(drugs),
total_pairs_evaluated=len(candidates),
top_candidates=top_candidates,
warnings=warnings,
)
save_cached_screening(request, response.model_dump())
return response
except HTTPException:
raise
except Exception as e:
logger.error(f"Error during screening: {str(e)}", exc_info=True)
raise HTTPException(status_code=500, detail=f"Screening failed: {str(e)}")
# Filter Results Endpoint
@app.get("/api/v1/results/potential")
async def get_potential_discoveries():
"""
Returns only potential drug discoveries (non-approved treatments).
This is a helper endpoint - use with stored results from screening.
"""
return {
"message": "Run screening first, then call this endpoint with results parameter"
}
# Error Handler
@app.exception_handler(HTTPException)
async def http_exception_handler(request: Request, exc: HTTPException):
"""Custom HTTP exception handler"""
return JSONResponse(
status_code=exc.status_code,
content={
"detail": exc.detail,
"status_code": exc.status_code
},
)
if __name__ == "__main__":
import uvicorn
uvicorn.run(
"app.main:app",
host=settings.HOST,
port=settings.PORT,
reload=settings.DEBUG
)
|