--- a
+++ b/app.py
@@ -0,0 +1,160 @@
+"""
+Streamlit app for demo.
+"""
+
+import streamlit as st
+from aitrika.engine.online_aitrika import OnlineAItrika
+from aitrika.utils.text_parser import generate_documents
+from aitrika.llm.openai import OpenAILLM
+from dotenv import load_dotenv
+import os
+import time
+
+
+load_dotenv()
+
+st.set_page_config(
+ page_title="AItrika",
+ page_icon="images/logo.png",
+)
+st.title("AItrika 🧪")
+
+
+def response_generator(query: str):
+ response = llm.query(query=query)
+ print(response)
+ for word in response.split():
+ yield word + " "
+ time.sleep(0.05)
+
+
+def reset_conversation():
+ st.session_state.conversation = None
+ st.session_state.chat_history = None
+
+
+def format_list_to_markdown(items):
+ markdown_text = ""
+ if items[0].strip().startswith("**") and items[0].strip().endswith("**"):
+ markdown_text += f"## {items[0].strip('* ')}\n\n"
+ items = items[1:]
+ for item in items:
+ parts = item.split("\n\n")
+ if len(parts) >= 2:
+ for subitem in parts[1:]:
+ markdown_text += f"- {subitem.strip()}\n"
+ else:
+ markdown_text += f"- {item.strip()}\n"
+ markdown_text = markdown_text.replace('"', "")
+ markdown_text += "\n"
+ return markdown_text
+
+
+pubmed_id = st.text_input("Enter the PubMed ID", placeholder="23747889")
+
+if pubmed_id:
+ engine = OnlineAItrika(pubmed_id=pubmed_id)
+ documents = generate_documents(content=engine.extract_abstract())
+ llm = OpenAILLM(documents=documents, api_key=os.getenv("GROQ_API_KEY"))
+ with st.expander("Select the information that you want to extract: "):
+ option = st.selectbox(
+ " ",
+ (
+ "PubMed ID",
+ "Title",
+ "Abstract",
+ "Authors",
+ "Full text",
+ "Journal",
+ "Genes",
+ "Diseases",
+ "Species",
+ "Chemicals",
+ "Mutations",
+ "Associations between genes and diseases",
+ "Results",
+ "Bibliography",
+ "Methods",
+ "Acknowledgements",
+ "Introduction",
+ "Paper results",
+ "Number of participants",
+ "Characteristics of participants",
+ ),
+ )
+ if option == "PubMed ID":
+ st.markdown("## PubMed ID")
+ st.write(engine.extract_pubmed_id())
+ elif option == "Title":
+ st.markdown("## Title")
+ st.write(engine.extract_title())
+ elif option == "Abstract":
+ st.markdown("## Abstract")
+ st.write(engine.extract_abstract())
+ elif option == "Authors":
+ st.markdown("## Authors")
+ st.write(engine.extract_authors())
+ elif option == "Full text":
+ st.markdown("## Full text")
+ st.write(engine.extract_full_text())
+ elif option == "Journal":
+ st.markdown("## Journal")
+ st.write(engine.extract_journal())
+ elif option == "Genes":
+ st.markdown("## Genes")
+ st.dataframe(engine.extract_genes(dataframe=True))
+ st.json(engine.genes())
+ elif option == "Diseases":
+ st.markdown("## Diseases")
+ st.dataframe(engine.extract_diseases(dataframe=True))
+ st.json(engine.diseases())
+ elif option == "Associations between genes and diseases":
+ st.markdown("## Associations between genes and diseases")
+ st.dataframe(engine.extract_associations(dataframe=True))
+ st.json(engine.extract_associations())
+ elif option == "Species":
+ st.markdown("## Species")
+ st.dataframe(engine.extract_species(dataframe=True))
+ st.json(engine.extract_species())
+ elif option == "Mutations":
+ st.markdown("## Mutations")
+ st.dataframe(engine.extract_mutations(dataframe=True))
+ st.json(engine.extract_mutations())
+ elif option == "Chemicals":
+ st.markdown("## Chemicals")
+ st.dataframe(engine.extract_chemicals(dataframe=True))
+ st.json(engine.extract_chemicals())
+ elif option == "Results":
+ st.markdown("## Results")
+ st.write(engine.extract_results(llm=llm))
+ elif option == "Bibliography":
+ st.markdown("## Bibliography")
+ st.write(engine.extract_bibliography(llm=llm))
+ elif option == "Methods":
+ st.markdown("## Methods")
+ methods = engine.extract_methods(llm=llm).split("---")
+ formatted_methods = format_list_to_markdown(methods)
+ st.markdown(formatted_methods)
+ elif option == "Acknowledgements":
+ st.markdown("## Acknowledgements")
+ st.write(engine.extract_acknowledgements(llm=llm))
+ elif option == "Introduction":
+ st.markdown("## Introduction")
+ st.write(engine.extract_introduction(llm=llm))
+ elif option == "Number of participants":
+ st.markdown("## Number of participants")
+ st.write(engine.extract_number_of_participants(llm=llm))
+ elif option == "Paper results":
+ st.markdown("## Paper results")
+ paper_results = engine.extract_paper_results(llm=llm).split("---")
+ formatted_paper_results = format_list_to_markdown(paper_results)
+ st.markdown(formatted_paper_results)
+ elif option == "Characteristics of participants":
+ st.markdown("## Characteristics of participants")
+ characteristics_of_participants = (
+ engine.extract_characteristics_of_participants(llm=llm).split("---")
+ )
+ formatted_characteristics_of_participants = format_list_to_markdown(
+ characteristics_of_participants
+ )
+ st.markdown(formatted_characteristics_of_participants)
