Spaces:
Runtime error
Runtime error
File size: 26,558 Bytes
69b6b11 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 |
"""
Streamlit application for the Pharmaceutical R&D Knowledge Ecosystem.
"""
import streamlit as st
import os
import pandas as pd
import json
import tempfile
import time
from datetime import datetime
from pdf_processor import PDFProcessor
from knowledge_store import KnowledgeStore
from llm_interface import LLMInterface
from graph_builder import (
init_handlers,
build_document_extraction_graph,
build_protocol_coach_graph,
build_content_authoring_graph,
build_traceability_graph
)
# =========================================================================
# App Setup and Configuration
# =========================================================================
# Page configuration
st.set_page_config(
page_title="Pharma R&D Knowledge Ecosystem",
page_icon="π",
layout="wide",
initial_sidebar_state="expanded"
)
# Initialize session state variables if they don't exist
if "chat_history" not in st.session_state:
st.session_state.chat_history = []
if "documents" not in st.session_state:
st.session_state.documents = []
if "knowledge_base_stats" not in st.session_state:
st.session_state.knowledge_base_stats = {
"documents": 0,
"studies": 0,
"endpoints": 0,
"objectives": 0,
"vectors": 0
}
# Initialize our handlers and graphs
@st.cache_resource
def initialize_app():
"""Initialize app resources and LangGraph workflows."""
# Get API key from environment or secrets
api_key = os.environ.get("ANTHROPIC_API_KEY")
if not api_key and hasattr(st, "secrets") and "ANTHROPIC_API_KEY" in st.secrets:
api_key = st.secrets["ANTHROPIC_API_KEY"]
# Initialize handlers
pdf_processor, knowledge_store, llm_interface = init_handlers(api_key)
# Build LangGraph workflows
extraction_graph = build_document_extraction_graph()
coach_graph = build_protocol_coach_graph()
authoring_graph = build_content_authoring_graph()
traceability_graph = build_traceability_graph()
return {
"pdf_processor": pdf_processor,
"knowledge_store": knowledge_store,
"llm_interface": llm_interface,
"extraction_graph": extraction_graph,
"coach_graph": coach_graph,
"authoring_graph": authoring_graph,
"traceability_graph": traceability_graph
}
# Initialize app resources
app_resources = initialize_app()
pdf_processor = app_resources["pdf_processor"]
knowledge_store = app_resources["knowledge_store"]
llm_interface = app_resources["llm_interface"]
extraction_graph = app_resources["extraction_graph"]
coach_graph = app_resources["coach_graph"]
authoring_graph = app_resources["authoring_graph"]
traceability_graph = app_resources["traceability_graph"]
# =========================================================================
# Helper Functions
# =========================================================================
def update_knowledge_base_stats():
"""Update the knowledge base statistics in session state."""
try:
# Get counts of different entity types
documents = knowledge_store.get_all_documents()
document_count = len(documents)
# Get unique protocol IDs
protocol_ids = set()
for doc in documents:
if "protocol_id" in doc and doc["protocol_id"]:
protocol_ids.add(doc["protocol_id"])
# Get vector store stats
vector_stats = knowledge_store.get_vector_store_stats()
vector_count = vector_stats.get("document_count", 0)
# Count objectives and endpoints across all protocols
objective_count = 0
endpoint_count = 0
for protocol_id in protocol_ids:
objectives = knowledge_store.get_objectives_by_protocol_id(protocol_id)
endpoints = knowledge_store.get_endpoints_by_protocol_id(protocol_id)
objective_count += len(objectives)
endpoint_count += len(endpoints)
# Update session state
st.session_state.knowledge_base_stats = {
"documents": document_count,
"studies": len(protocol_ids),
"objectives": objective_count,
"endpoints": endpoint_count,
"vectors": vector_count
}
except Exception as e:
st.error(f"Error updating knowledge base stats: {e}")
def process_document(uploaded_file):
"""Process an uploaded document and store in knowledge base."""
try:
# Create a progress bar
progress_bar = st.progress(0)
status_text = st.empty()
# Step 1: Save the uploaded file
status_text.text("Saving uploaded file...")
progress_bar.progress(10)
# Save uploaded file temporarily
with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp_file:
tmp_file.write(uploaded_file.getbuffer())
file_path = tmp_file.name
# Step 2: Process through LangGraph extraction workflow
status_text.text("Parsing document...")
progress_bar.progress(20)
# Initialize state for extraction
initial_state = {
"document_path": file_path,
"status": "initialized"
}
# Run extraction workflow
result_state = extraction_graph.invoke(initial_state)
# Update progress based on status
if result_state.get("status") == "error":
status_text.text(f"Error: {result_state.get('error', 'Unknown error')}")
progress_bar.progress(100)
return {
"status": "error",
"error": result_state.get("error", "Unknown error"),
"filename": uploaded_file.name
}
# Update progress
status_text.text("Processing completed successfully!")
progress_bar.progress(100)
# Update knowledge base stats
update_knowledge_base_stats()
# Return result
return {
"status": "success",
"filename": uploaded_file.name,
"document_id": result_state.get("document_id"),
"protocol_id": result_state.get("extracted_study", {}).get("protocol_id")
}
except Exception as e:
st.error(f"Error processing document: {e}")
return {
"status": "error",
"error": str(e),
"filename": uploaded_file.name
}
finally:
# Clean up temporary file
if 'file_path' in locals():
try:
os.unlink(file_path)
except:
pass
def chat_with_protocol_coach(query):
"""Process a query through the Protocol Coach."""
try:
# Initialize state for Protocol Coach
initial_state = {
"query": query,
"chat_history": st.session_state.chat_history
}
# Run Protocol Coach workflow
result_state = coach_graph.invoke(initial_state)
return {
"status": "success",
"response": result_state.get("response", "I couldn't generate a response."),
"context": result_state.get("retrieved_context", [])
}
except Exception as e:
return {
"status": "error",
"error": str(e)
}
def generate_document_section(section_type, protocol_id=None, style_guide=None):
"""Generate a document section using the content authoring workflow."""
try:
# Initialize state for Content Authoring
initial_state = {
"section_type": section_type,
"target_protocol_id": protocol_id,
"style_guide": style_guide
}
# Run Content Authoring workflow
result_state = authoring_graph.invoke(initial_state)
return {
"status": "success",
"content": result_state.get("generated_content", "I couldn't generate the content."),
"context": result_state.get("retrieved_context", [])
}
except Exception as e:
return {
"status": "error",
"error": str(e)
}
def analyze_document_traceability(source_id, target_id, entity_type):
"""Analyze traceability between two documents."""
try:
# Initialize state for Traceability Analysis
initial_state = {
"source_document_id": source_id,
"target_document_id": target_id,
"entity_type": entity_type
}
# Run Traceability Analysis workflow
result_state = traceability_graph.invoke(initial_state)
return {
"status": "success",
"analysis": result_state.get("analysis", "I couldn't perform the analysis."),
"matched_pairs": result_state.get("matched_pairs", [])
}
except Exception as e:
return {
"status": "error",
"error": str(e)
}
# =========================================================================
# Sidebar: Document Upload and Management
# =========================================================================
def render_sidebar():
"""Render the sidebar for document management."""
st.sidebar.title("Document Management")
# Knowledge Base Stats
st.sidebar.subheader("Knowledge Base Stats")
stats = st.session_state.knowledge_base_stats
col1, col2 = st.sidebar.columns(2)
col1.metric("Documents", stats["documents"])
col2.metric("Studies", stats["studies"])
col1.metric("Objectives", stats["objectives"])
col2.metric("Endpoints", stats["endpoints"])
st.sidebar.metric("Vector Chunks", stats["vectors"])
# Document Upload
st.sidebar.subheader("Upload Documents")
uploaded_files = st.sidebar.file_uploader(
"Upload Protocol/SAP PDFs",
type="pdf",
accept_multiple_files=True,
help="Upload clinical documents (Protocol, SAP, etc.) to add to the knowledge base."
)
# Process uploaded files if any
if uploaded_files:
if st.sidebar.button("Process Documents"):
with st.sidebar.expander("Processing Results", expanded=True):
for uploaded_file in uploaded_files:
st.write(f"Processing: {uploaded_file.name}")
result = process_document(uploaded_file)
if result["status"] == "success":
st.success(f"Successfully processed {result['filename']}")
# Add to documents list if not already there
doc_exists = False
for doc in st.session_state.documents:
if doc.get("filename") == result["filename"]:
doc_exists = True
break
if not doc_exists:
st.session_state.documents.append({
"filename": result["filename"],
"document_id": result.get("document_id"),
"protocol_id": result.get("protocol_id"),
"processed_date": datetime.now().strftime("%Y-%m-%d %H:%M")
})
else:
st.error(f"Error processing {result['filename']}: {result.get('error', 'Unknown error')}")
# Document list
st.sidebar.subheader("Processed Documents")
if not st.session_state.documents:
st.sidebar.info("No documents processed yet.")
else:
for i, doc in enumerate(st.session_state.documents):
with st.sidebar.expander(f"{doc['filename']}"):
st.write(f"**Protocol ID:** {doc.get('protocol_id', 'Unknown')}")
st.write(f"**Processed:** {doc.get('processed_date', 'Unknown')}")
# Refresh Stats Button
if st.sidebar.button("Refresh Stats"):
update_knowledge_base_stats()
st.sidebar.success("Stats refreshed!")
# =========================================================================
# Main Content Tabs
# =========================================================================
def render_protocol_coach_tab():
"""Render the Protocol Coach chatbot tab."""
st.header("Protocol Coach Chatbot")
st.info("Ask questions about the protocol documents in the knowledge base. The Protocol Coach will retrieve relevant information to answer your questions.")
# Initialize or display chat history
for message in st.session_state.chat_history:
with st.chat_message(message["role"]):
st.markdown(message["content"])
# Chat input
if query := st.chat_input("Ask about protocols..."):
# Add user message to chat history and display
st.session_state.chat_history.append({"role": "user", "content": query})
with st.chat_message("user"):
st.markdown(query)
# Process query
with st.chat_message("assistant"):
with st.spinner("Thinking..."):
result = chat_with_protocol_coach(query)
if result["status"] == "success":
st.markdown(result["response"])
# Show context sources if debug mode enabled
if st.session_state.get("debug_mode", False):
with st.expander("Context Sources"):
for i, ctx in enumerate(result.get("context", [])):
st.write(f"**Source {i+1}:** {ctx.get('metadata', {}).get('source', 'Unknown')}")
st.write(f"**Section:** {ctx.get('metadata', {}).get('section', 'Unknown')}")
st.write("---")
# Add assistant response to chat history
st.session_state.chat_history.append({"role": "assistant", "content": result["response"]})
else:
st.error(f"Error: {result.get('error', 'Unknown error')}")
st.session_state.chat_history.append({"role": "assistant", "content": f"Error: {result.get('error', 'Unknown error')}"})
def render_content_authoring_tab():
"""Render the Content Authoring tab."""
st.header("Content Authoring Assistant")
st.info("Generate document sections based on knowledge extracted from similar documents.")
col1, col2 = st.columns([1, 1])
with col1:
st.subheader("Content Generation Settings")
# Section Type Selection
section_types = [
"Introduction",
"Objectives and Endpoints",
"Study Design",
"Study Population",
"Statistical Considerations",
"Inclusion Criteria",
"Exclusion Criteria",
"Safety Assessments",
"Pharmacokinetic Assessments"
]
section_type = st.selectbox("Select Section Type", section_types)
# Protocol Selection for Context (Optional)
protocol_options = ["--None--"]
for doc in st.session_state.documents:
if doc.get("protocol_id"):
protocol_options.append(doc.get("protocol_id"))
target_protocol = st.selectbox(
"Target Protocol ID (Optional)",
protocol_options
)
target_protocol = None if target_protocol == "--None--" else target_protocol
# Style Guide (Optional)
style_guide = st.text_area(
"Style Guide (Optional)",
placeholder="Enter any specific style guidelines or content requirements..."
)
# Generate Button
generate_button = st.button("Generate Content")
# Debug toggle
st.session_state.debug_mode = st.checkbox("Show Context Sources", value=st.session_state.get("debug_mode", False))
with col2:
st.subheader("Generated Content")
if generate_button:
with st.spinner("Generating content..."):
result = generate_document_section(
section_type=section_type,
protocol_id=target_protocol,
style_guide=style_guide if style_guide else None
)
if result["status"] == "success":
st.markdown(result["content"])
# Show context sources if debug mode enabled
if st.session_state.get("debug_mode", False):
with st.expander("Context Sources"):
for i, ctx in enumerate(result.get("context", [])):
st.write(f"**Source {i+1}:** {ctx.get('metadata', {}).get('source', 'Unknown')}")
st.write(f"**Section:** {ctx.get('metadata', {}).get('section', 'Unknown')}")
st.write("---")
else:
st.error(f"Error: {result.get('error', 'Unknown error')}")
def render_traceability_tab():
"""Render the Document Traceability tab."""
st.header("Cross-Document Traceability")
st.info("Analyze relationships between related documents (e.g., Protocol and SAP).")
col1, col2 = st.columns([1, 1])
with col1:
st.subheader("Traceability Analysis Settings")
# Document Selection
document_options = []
for doc in st.session_state.documents:
document_options.append({
"id": doc.get("document_id", ""),
"label": f"{doc['filename']} ({doc.get('protocol_id', 'Unknown')})"
})
# Source Document
source_options = [{"id": "", "label": "--Select Source Document--"}] + document_options
source_doc = st.selectbox(
"Source Document",
options=source_options,
format_func=lambda x: x["label"]
)
# Target Document
target_options = [{"id": "", "label": "--Select Target Document--"}] + document_options
target_doc = st.selectbox(
"Target Document",
options=target_options,
format_func=lambda x: x["label"]
)
# Entity Type
entity_types = [
{"id": "objectives", "label": "Study Objectives"},
{"id": "endpoints", "label": "Endpoints"},
{"id": "population", "label": "Population Criteria"}
]
entity_type = st.selectbox(
"Entity Type to Compare",
options=entity_types,
format_func=lambda x: x["label"]
)
# Analyze Button
analyze_button = st.button("Analyze Traceability")
with col2:
st.subheader("Analysis Results")
if analyze_button:
if not source_doc["id"] or not target_doc["id"]:
st.error("Please select both source and target documents.")
else:
with st.spinner("Analyzing traceability..."):
result = analyze_document_traceability(
source_id=source_doc["id"],
target_id=target_doc["id"],
entity_type=entity_type["id"]
)
if result["status"] == "success":
st.markdown(result["analysis"])
# Show matched pairs if debug mode enabled
if st.session_state.get("debug_mode", False) and result.get("matched_pairs"):
with st.expander("Matched Entity Pairs"):
for i, pair in enumerate(result["matched_pairs"]):
st.write(f"**Pair {i+1}**")
st.write(f"**Source:** {pair.get('source_text', 'Unknown')}")
st.write(f"**Target:** {pair.get('target_text', 'Unknown')}")
st.write("---")
else:
st.error(f"Error: {result.get('error', 'Unknown error')}")
def render_knowledge_explorer_tab():
"""Render the Knowledge Base Explorer tab."""
st.header("Knowledge Base Explorer")
st.info("Explore the structured data extracted from documents in the knowledge base.")
# Entity Type Selection
entity_types = [
{"id": "studies", "label": "Studies"},
{"id": "objectives", "label": "Study Objectives"},
{"id": "endpoints", "label": "Endpoints"},
{"id": "population", "label": "Population Criteria"},
{"id": "documents", "label": "Documents"}
]
entity_type = st.selectbox(
"Select Entity Type",
options=entity_types,
format_func=lambda x: x["label"]
)
# Filter by Protocol ID (Optional)
protocol_options = ["--All Protocols--"]
for doc in st.session_state.documents:
if doc.get("protocol_id") and doc.get("protocol_id") not in protocol_options:
protocol_options.append(doc.get("protocol_id"))
filter_protocol = st.selectbox(
"Filter by Protocol ID",
protocol_options
)
filter_protocol = None if filter_protocol == "--All Protocols--" else filter_protocol
# Search Query (Optional)
search_query = st.text_input(
"Search Query (Optional)",
placeholder="Enter text to search for..."
)
# Display Results
st.subheader("Results")
try:
# Retrieve data based on entity type
if entity_type["id"] == "studies":
if filter_protocol:
data = [knowledge_store.get_study_by_protocol_id(filter_protocol)]
else:
data = knowledge_store.get_all_studies()
elif entity_type["id"] == "objectives":
if filter_protocol:
data = knowledge_store.get_objectives_by_protocol_id(filter_protocol)
else:
# Get all objectives across protocols
data = []
documents = knowledge_store.get_all_documents()
protocol_ids = set()
for doc in documents:
if "protocol_id" in doc and doc["protocol_id"]:
protocol_ids.add(doc["protocol_id"])
for pid in protocol_ids:
data.extend(knowledge_store.get_objectives_by_protocol_id(pid))
elif entity_type["id"] == "endpoints":
if filter_protocol:
data = knowledge_store.get_endpoints_by_protocol_id(filter_protocol)
else:
# Get all endpoints across protocols
data = []
documents = knowledge_store.get_all_documents()
protocol_ids = set()
for doc in documents:
if "protocol_id" in doc and doc["protocol_id"]:
protocol_ids.add(doc["protocol_id"])
for pid in protocol_ids:
data.extend(knowledge_store.get_endpoints_by_protocol_id(pid))
elif entity_type["id"] == "population":
if filter_protocol:
data = knowledge_store.get_population_criteria_by_protocol_id(filter_protocol)
else:
# Get all population criteria across protocols
data = []
documents = knowledge_store.get_all_documents()
protocol_ids = set()
for doc in documents:
if "protocol_id" in doc and doc["protocol_id"]:
protocol_ids.add(doc["protocol_id"])
for pid in protocol_ids:
data.extend(knowledge_store.get_population_criteria_by_protocol_id(pid))
elif entity_type["id"] == "documents":
if filter_protocol:
data = knowledge_store.get_documents_by_protocol_id(filter_protocol)
else:
data = knowledge_store.get_all_documents()
else:
data = []
# Filter by search query if provided
if search_query:
filtered_data = []
search_lower = search_query.lower()
for item in data:
# Convert item to string for searching
item_str = json.dumps(item).lower()
if search_lower in item_str:
filtered_data.append(item)
data = filtered_data
# Display results
if not data:
st.info("No data found.")
else:
st.write(f"{len(data)} items found")
# Display as table if possible, otherwise as JSON
try:
df = pd.DataFrame(data)
st.dataframe(df, use_container_width=True)
except Exception as e:
st.json(data)
except Exception as e:
st.error(f"Error retrieving data: {e}")
# =========================================================================
# Main App
# =========================================================================
def main():
"""Main application function."""
st.title("π§ Pharmaceutical R&D Knowledge Ecosystem")
# Render the sidebar for document management
render_sidebar()
# Initialize knowledge base stats on first load
if st.session_state.knowledge_base_stats["documents"] == 0:
update_knowledge_base_stats()
# Main content tabs
tab1, tab2, tab3, tab4 = st.tabs([
"π Content Authoring",
"π€ Protocol Coach",
"π Knowledge Explorer",
"π Cross-Document Traceability"
])
with tab1:
render_content_authoring_tab()
with tab2:
render_protocol_coach_tab()
with tab3:
render_knowledge_explorer_tab()
with tab4:
render_traceability_tab()
# Footer
st.markdown("---")
st.caption("Pharmaceutical R&D Knowledge Ecosystem | A demonstration of AI-assisted document processing and knowledge management")
if __name__ == "__main__":
main() |