License		Python
Package		Build status
Tests	Coverage coming soon.	Docker
Development		Contributions

Generative AI models have shown tremendous usefulness in increasing accessibility and automation of a wide range of tasks. Yet, their application to the biomedical domain is still limited, in part due to the lack of a common framework for deploying, testing, and evaluating the diverse models and auxiliary technologies that are needed. This repository contains the biochatter Python package, a generic backend library for the connection of biomedical applications to conversational AI. Described in this preprint and used in ChatGSE, which is being developed at https://github.com/biocypher/ChatGSE. More to come, so stay tuned!

BioChatter is part of the BioCypher ecosystem, connecting natively to BioCypher knowledge graphs. The BioChatter paper is being written here.

To use the package, install it from PyPI, for instance using pip (pip install biochatter) or Poetry (poetry add biochatter).

The package has some optional dependencies that can be installed using the following extras (e.g. pip install biochatter[xinference]):

• xinference: support for querying open-source LLMs through Xorbits Inference

• podcast: support for podcast text-to-speech (for the free Google TTS; the paid OpenAI TTS can be used without this extra)

• streamlit: support for streamlit UI functions (used in ChatGSE)

Check out the documentation for examples, use cases, and more information. Many common functionalities covered by BioChatter can be seen in use in the ChatGSE code base.

Check out this repository for more info on computational biology usage of large language models.

Due to some incompatibilities of pymilvus with Apple Silicon, we have created a dev container for this project. To use it, you need to have Docker installed on your machine. Then, you can run the devcontainer setup as recommended by VSCode here or using Docker directly.

The dev container expects an environment file (there are options, but the basic one is .devcontainer/local.env) with the following variables:

OPENAI_API_KEY=(sk-...)
DOCKER_COMPOSE=true
DEVCONTAINER=true

To test vector database functionality, you also need to start a Milvus standalone server. You can do this by running docker-compose up as described here on the host machine (not from inside the devcontainer).