Skip to content

Home

Talk2BioModels Talk2Cells Talk2KnowledgeGraphs TESTS Talk2Scholars TESTS Talk2AIAgents4Pharma GitHub Release Python Version from PEP 621 TOML Docker Pulls

Introduction

Welcome to AIAgents4Pharma – an open-source project by Team VPE that brings together AI-driven tools to help researchers and pharma interact seamlessly with complex biological data.

Our toolkit currently consists of the following agents:

  • Talk2BioModels (v1 released; v2 in progress): Engage directly with mathematical models in systems biology.
  • Talk2KnowledgeGraphs (v1 in progress): Access and explore complex biological knowledge graphs for insightful data connections.
  • Talk2Scholars (v1 in progress): Get recommendations for articles related to your choice. Download, query, and write/retrieve them to your reference manager (currently supporting Zotero).
  • Talk2Cells (v1 in progress): Query and analyze sequencing data with ease.
  • Talk2AIAgents4Pharma (v1 in progress): Converse with all the agents above (currently supports T2B and T2KG)

AIAgents4Pharma

Getting Started

Python Version from PEP 621 TOML

Installation

Please use version 1.26.2 or later for better support with NVIDIA NIM models.

Option 1: PyPI

Python Version from PEP 621 TOML

pip install aiagents4pharma

Check out the tutorials on each agent for detailed instructions.

Option 2: Docker Hub

We now have Talk2AIAgents4Pharma, Talk2Biomodels, and Talk2Scholars available on Docker Hub.

Run Talk2AIAgents4Pharma and Talk2KnowledgeGraphs

Talk2AIAgents4Pharma and Talk2KnowledgeGraphs require Ollama for embedding models, so Docker Compose is used to run both containers in the same network.

Setup Environment Variables
  1. Choose the app you want to use:
# Navigate to the correct directory before setting up environment variables.
# Use one of the following commands based on the agent you want to use:
cd AIAgents4Pharma/aiagents4pharma/talk2aiagents4pharma
cd AIAgents4Pharma/aiagents4pharma/talk2knowledgegraphs
  1. Copy the .env.example file and rename it to .env: 
    cp .env.example .env
  2. Open the .env file and add your API keys:
OPENAI_API_KEY=your_openai_api_key
NVIDIA_API_KEY=your_nvidia_api_key
OLLAMA_HOST=http://ollama:11434
LANGCHAIN_TRACING_V2=true
LANGCHAIN_API_KEY=your_langchain_api_key_here
# Notes:
# The API endpoint for Ollama is already set in env.example.
# Both API keys (OPENAI_API_KEY and NVIDIA_API_KEY) are required for Talk2AIAgents4Pharma.
# If using Talk2KnowledgeGraphs separately, only the OPENAI_API_KEY is needed.
# Langsmith API for tracing is optional for both, set it in env.example if required.
  1. Save the file.

To start the containers, run the following command:

docker compose --profile cpu up # for CPU mode
docker compose --profile nvidia up # for GPU mode
docker compose --profile amd up # for AMD mode

This will:

  • Pull the latest images if they are not already available.
  • Start both Talk2AIAgents4Pharma or Talk2KnowledgeGraphs and Ollama containers in the same network.
  • Ensure Ollama is running first before launching Talk2AIAgents4Pharma or Talk2KnowledgeGraphs.

To Access the web app, open your browser and go to:

http://localhost:8501

To stop the containers, run:

docker compose down
Run Talk2Biomodels and Talk2Scholars
  1. Run the containers
docker run -d \
  --name talk2biomodels \
  -e OPENAI_API_KEY=<your_openai_api_key> \
  -e NVIDIA_API_KEY=<your_nvidia_api_key> \
  -p 8501:8501 \
  virtualpatientengine/talk2biomodels

docker run -d \
  --name talk2scholars \
  -e OPENAI_API_KEY=<your_openai_api_key> \
  -e ZOTERO_API_KEY=<your_zotero_api_key> \
  -e ZOTERO_USER_ID=<your_zotero_user_id> \
  -p 8501:8501 \
  virtualpatientengine/talk2scholars
  1. Access the Web App Open your browser and go to:
http://localhost:8501

You can create a free account at NVIDIA and apply for their free credits here.

Notes

  • Ensure you replace <your_openai_api_key>, <your_nvidia_api_key>, <your_zotero_api_key>, and <your_zotero_user_id> with your actual credentials.
  • Both applications use port 8501, so run them on different ports if needed: 
    docker run -d -e OPENAI_API_KEY=<your_openai_api_key> -p 8501:8501 virtualpatientengine/talk2scholars
    Then, access it via http://localhost:8501.

Option 3: git

Python Version from PEP 621 TOML

  1. Clone the repository: 
    git clone https://github.com/VirtualPatientEngine/AIAgents4Pharma
cd AIAgents4Pharma
  2. Install dependencies:
pip install -r requirements.txt

⚠️ The current version of T2KG requires additional Ollama library to be installed.

Ollama can be easily downloaded and installed from the following link: https://ollama.com/download

As an alternative, use the following commands to install the library using terminal and to pull necessary model:

  • Ubuntu: 
    curl -fsSL https://ollama.com/install.sh | sh
ollama pull nomic-embed-text
  • Windows: 
    curl -L https://ollama.com/download/ollama-windows-amd64.zip -o ollama-windows-amd64.zip
tar -xzf .\ollama-windows-amd64.zip
start ollama serve
ollama pull nomic-embed-text
  • macOS: 
    brew install ollama
ollama pull nomic-embed-text
    A list of pulled Ollama models can be checked using the following command:
ollama list

⚠️ pcst_fast 1.0.10 library requires Microsoft Visual C++ 14.0 or greater to be installed. You can download Microsoft C++ Build Tools from here.

  1. Initialize OPENAI_API_KEY and NVIDIA_API_KEY
export OPENAI_API_KEY=....
export NVIDIA_API_KEY=....

You can create a free account at NVIDIA and apply for their free credits here.

  1. Initialize ZOTERO_API_KEY and ZOTERO_USER_ID
export ZOTERO_API_KEY=....
export ZOTERO_USER_ID=....

Please note that ZOTERO keys are requried only if you want to launch Talk2Scholars. For all the other agents, please ignore this step.

  1. [Optional] Initialize LANGSMITH_API_KEY
export LANGCHAIN_TRACING_V2=true
export LANGCHAIN_API_KEY=<your-api-key>

Please note that this will create a new tracing project in your Langsmith account with the name T2X-xxxx, where X can be AA4P (Main Agent), B (Biomodels), S (Scholars), KG (KnowledgeGraphs), or C (Cells). If you skip the previous step, it will default to the name default. xxxx will be the 4-digit ID created for the session.

  1. Launch the app: 
    streamlit run app/frontend/streamlit_app_<agent>.py
    Replace with the agent name you are interested to launch.

For detailed instructions on each agent, please refer to their respective modules.

Contributing

We welcome contributions to AIAgents4Pharma! Here’s how you can help:

  1. Fork the repository
  2. Create a new branch for your feature (git checkout -b feat/feature-name)
  3. Commit your changes (git commit -m 'feat: Add new feature')
  4. Push to the branch (git push origin feat/feature-name)
  5. Open a pull request and reach out to any one of us below via Discussions:

Note: We welcome all contributions, not just programming-related ones. Feel free to open bug reports, suggest new features, or participate as a beta tester. Your support is greatly appreciated!

Current Needs

  • Beta testers for Talk2BioModels and Talk2Scholars.
  • Developers with experience in Python and Bioinformatics and/or knowledge graphs for contributions to AIAgents4Pharma.

Feel free to reach out to us via Discussions.

Check out our CONTRIBUTING.md for more information.

Feedback

Questions/Bug reports/Feature requests/Comments/Suggestions? We welcome all. Please use Issues or Discussions 😀