Pubmed ChatGPT

Pubmed_ChatGPT.py

import sys
import openai
import json
import pandas as pd
from Bio import Entrez
from datetime import datetime
from PyQt6 import QtWidgets, QtGui
from PyQt6.QtCore import Qt

# Replace with your own ChatGPT API key
openai.api_key = "API Key"
Entrez.email = "Email"  # Replace with your email

def generate_overall_introduction(article_infos):
    prompt = f"Write an overall introduction for a research paper based on the following articles:\n\n{article_infos}\n\nCite each article in the introduction and provide a reference list at the end."
    
    with open("prompt.txt", "w") as f:
        f.write(prompt)
    
    response = openai.ChatCompletion.create(
        model="gpt-3.5-turbo",
        messages=[{"role": "assistant", "content": prompt}]
    )
    return response["choices"][0]["message"]["content"]


def search_pubmed_articles(keywords, max_results=5):
    search_term = " AND ".join(keywords)
    handle = Entrez.esearch(db="pubmed", term=search_term, sort="relevance", retmax=max_results)
    record = Entrez.read(handle)
    handle.close()
    return record["IdList"]

def fetch_article_details(article_id):
    handle = Entrez.efetch(db="pubmed", id=article_id, rettype="xml", retmode="text")
    article = Entrez.read(handle)["PubmedArticle"][0]
    handle.close()
    return article

def format_authors(authors):
    formatted_authors = []
    for author in authors:
        last_name = author.get("LastName", "")
        initials = author.get("Initials", "")
        formatted_authors.append(f"{last_name}, {initials}.")
    return ", ".join(formatted_authors)

def summarize_abstract(abstract):
    prompt = "Summarize the following abstract in three bullet points:"+abstract
    response = openai.ChatCompletion.create(
        model="gpt-3.5-turbo",
        messages=[{"role": "assistant", "content":prompt}]
    )
    return response["choices"][0]["message"]["content"]

def format_apa_citation(article):
    authors = format_authors(article["MedlineCitation"]["Article"]["AuthorList"])
    title = article["MedlineCitation"]["Article"]["ArticleTitle"]
    journal = article["MedlineCitation"]["Article"]["Journal"]["Title"]
    pub_year = datetime.strptime(article["MedlineCitation"]["Article"]["Journal"]["JournalIssue"]["PubDate"]["Year"], "%Y").year
    volume = article["MedlineCitation"]["Article"]["Journal"]["JournalIssue"]["Volume"]
    #issue = article["MedlineCitation"]["Article"]["Journal"]["JournalIssue"]["Issue"]
    pages = article["MedlineCitation"]["Article"]["Pagination"]["MedlinePgn"]

    return f"{authors} ({pub_year}). {title}. {journal}, {volume}, {pages}."



class MainWindow(QtWidgets.QWidget):
    def __init__(self):
        super().__init__()

        self.init_ui()

    def init_ui(self):
        self.keywords_label = QtWidgets.QLabel('Keywords:')
        self.keywords_input = QtWidgets.QLineEdit()
        self.search_button = QtWidgets.QPushButton('Search')
        self.search_button.clicked.connect(self.search_and_export)
        self.status_label = QtWidgets.QLabel('')

        self.layout = QtWidgets.QVBoxLayout()
        self.layout.addWidget(self.keywords_label)
        self.layout.addWidget(self.keywords_input)
        self.layout.addWidget(self.search_button)
        self.layout.addWidget(self.status_label)

        self.setLayout(self.layout)
        self.setWindowTitle('PubMed Article Search')


    def generate_overall_introduction(article_infos):
        prompt = f"Write an overall introduction for a research paper based on the following articles:\n\n{article_infos}\n\nCite each article in the introduction and provide a reference list at the end."
        
        with open("prompt.txt", "w") as f:
            f.write(prompt)
        
        response = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",
            messages=[{"role": "assistant", "content": prompt}]
        )
        return response["choices"][0]["message"]["content"]

    # ... (Keep the MainWindow class and init_ui function)

    def search_and_export(self):
        self.status_label.setText('Searching...')
        QtWidgets.QApplication.processEvents()

        keywords = self.keywords_input.text().split(', ')
        article_ids = search_pubmed_articles(keywords)

        results = []
        article_infos = ""
        for i, article_id in enumerate(article_ids):
            article = fetch_article_details(article_id)
            authors = format_authors(article["MedlineCitation"]["Article"]["AuthorList"])
            title = article["MedlineCitation"]["Article"]["ArticleTitle"]
            journal = article["MedlineCitation"]["Article"]["Journal"]["Title"]
            abstract = article["MedlineCitation"]["Article"]["Abstract"]["AbstractText"][0]
            pub_year = datetime.strptime(article["MedlineCitation"]["Article"]["Journal"]["JournalIssue"]["PubDate"]["Year"], "%Y").year
            apa_citation = format_apa_citation(article)

            results.append({
                "Authors": authors,
                "Title": title,
                "Journal": journal,
                "Publication Year": pub_year,
                "Abstract": abstract,
                "APA Citation": apa_citation
            })

            article_info = f"Article {i+1}:\nAuthors: {authors}\nTitle: {title}\nJournal: {journal}\nPublication Year: {pub_year}\nAbstract: {abstract}\nAPA Citation: {apa_citation}\n\n"
            article_infos += article_info

        # Save results to Excel file
        df = pd.DataFrame(results)
        df.to_excel("pubmed_results.xlsx", index=False)

        overall_introduction = generate_overall_introduction(article_infos)

        with open("overall_introduction.txt", "w") as f:
            f.write(overall_introduction)

        self.status_label.setText('Done!')


if __name__ == '__main__':
    app = QtWidgets.QApplication(sys.argv)
    main_window = MainWindow()
    main_window.show()
    sys.exit(app.exec())