NagGPT/GFA

NagGPT logo

GFA (Genomics Fetcher-Analyzer) is an OpenAI custom GPT that enables ChatGPT to perform some bioinformatics database lookups and analyses. To converse with GFA from chatgpt.com, users must provide a shared Google Drive folder that gives anyone with the link editor permission.

Setup and use instructions

  1. Set up a Google Drive folder (detailed instructions below).
  2. Start a conversation with GFA, providing your folder's ID at the beginning of the conversation.
  3. When prompted by GFA to make files available, attach them as follows:
    • A. Click the "+" button ("Attach file") to the left of the prompt. Select "Add from Google Drive" from the popup menu, as shown in the following screenshot:
      Screenshot of ChatGPT GUI to attach file
    • B. Select the files requested by GFA within your shared Google Drive folder

Important

Users should make sure to:

  1. Give GFA their Google Drive ID at the beginning of each conversation, even if GFA does not ask for it.
  2. Redirect GFA as necessary in the course of the conversation.
  3. Keep in mind that GFA will make errors in the course of conversations
  4. Refrain from using GFA for anything important

How to Create a Google Drive Folder

Prerequisites

Step 1: Create a New Google Drive Folder

  1. Open Google Drive and sign in.
  2. Click on the "+ New" button and select "Folder".
  3. Name your folder (e.g., "GFA_data_folder").

Step 2: Make the Folder Editable to Anyone

  1. Right-click on the folder and select "Share".
  2. Click on "General access" and select "Anyone with the link".
  3. Change the permission to "Editor".
  4. Click "Copy link" and save it for reference.

Step 3: Retrieve the Folder ID

The folder ID is found in the copied link:

https://drive.google.com/drive/folders/1aBcD2EfG3HIj4KlmNoPqRsTuVwXyZ

The folder ID is the unique string of characters following /folders/. In this example, the folder ID is 1aBcD2EfG3HIj4KlmNoPqRsTuVwXyZ. Copy this ID to use later.


Reference

Cinquin O. Steering veridical Large Language Model analyses by correcting and enriching generated database queries: first steps toward ChatGPT bioinformatics. Briefings in Bioinformatics (2025). https://doi.org/10.1093/bib/bbaf045