Sovereign 8s.is mirror of standardgalactic/LLM-prioritization-Framework
  • Jupyter Notebook 100%
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2025-12-15 17:08:01 -06:00
result112025 Add files via upload 2025-12-15 17:08:01 -06:00
.DS_Store Initial commit with data and code 2025-09-02 15:08:16 -05:00
0623LLM-based priotization strategy.png Initial commit with data and code 2025-09-02 15:08:16 -05:00
AD LLM Evaluation.png Initial commit with data and code 2025-09-02 15:08:16 -05:00
drug_disease_with_numbered_abstracts_20250325_190703.csv Add files via upload 2025-12-15 16:57:40 -06:00
LLM_validation1125.ipynb Add files via upload 2025-12-15 16:57:40 -06:00
README.md Revise input files section in README 2025-12-15 17:00:17 -06:00
resultanalysis.ipynb Add files via upload 2025-12-15 16:57:40 -06:00

LLM-based Prioritization Framework for Alzheimer’s Therapeutics

This repository contains the LLM-driven evidence synthesis pipeline and the replication guide for the paper: "Bridging the Computational-Experimental Gap: Leveraging Large Language Model to Prioritize Alzheimer’s Therapeutics Based on Comparison of Learning Models".

📂 Repository Contents

  • LLM_validation1125run.ipynb: The master notebook that executes the core pipeline:

    1. Data Loading: Imports the 90 candidate drugs generated by the graph models and the corresponding PubMed abstracts.
    2. Evidence Synthesis: Queries the GPT-4.1 API.
  • resultanalysis.ipynb: Applies the prioritization criteria to rank the drugs based on the LLM outputs.

  • drug_disease_long_format_20250325_190703.csv: Contains the input files.

🚀 How to Run the Code

The workflow is consolidated into the LLM_validation1125run.ipynb notebook for ease of replication.

  1. Set your OpenAI API key in a file named openai_api.txt.
  2. Open LLM_validation1125run.ipynb.
  3. Run all cells to execute the evidence synthesis.
  4. Run resultanalysis.ipynb to generate the final lists.

🔬 Replication Guide for Candidate Generation (Graph Models)

The input for LLM_validation1125run.ipynb is a list of candidate drugs. These candidates were generated using three state-of-the-art models. To replicate this upstream step, we utilized the official implementations adapted for the PrimeKG dataset:

1. TxGNN

We utilized the official pre-trained weights and zero-shot prediction scores.

2. CompGCN

We retrained the model on PrimeKG using the standard link prediction setting.

3. RLR (Hetionet)

We computed Degree-Weighted Path Counts (DWPC) using the learn and hetnet_ml packages.