MDCrow is an LLM-agent based toolset for Molecular Dynamics. It's built using Langchain and uses a collection of tools to set up and execute molecular dynamics simulations, particularly in OpenMM.
To use the OpenMM features in the agent, please set up a conda environment following these steps.
conda env create -n mdcrow -f environment.yaml
conda activate mdcrow
If you already have a conda environment, you can install dependencies before you activate it with the following steps.
- Install the necessary conda dependencies:
conda env update -n <YOUR_CONDA_ENV_HERE> -f environment.yaml
pip install git+https://github.com/ur-whitelab/MDCrow.git
The next step is to set up your API keys in your environment. An API key for an LLM provider is necessary for this project. Supported LLM providers are OpenAI, TogetherAI, Fireworks, and Anthropic. We recommend setting up the API keys in a .env file. You can use the provided .env.example file as a template.
- Copy the
.env.examplefile and rename it to.env:cp .env.example .env - Replace the placeholder values in
.envwith your actual keys
You can ask MDCrow to conduct molecular dynamics tasks using OpenAI's GPT model.
from mdcrow import MDCrow
agent = MDCrow(model="gpt-3.5-turbo")
agent.run("Simulate protein 1ZNI at 300 K for 0.1 ps and calculate the RMSD over time.")
Note: To distinguish Together models from the rest, you'll need to add the "together" prefix in the model flag, such as agent = MDCrow(model="together/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo")
By default, we support LLMs through OpenAI API. However, feel free to use other LLM providers. Make sure to install the necessary package for it. Here's a list of packages required for alternative LLM providers we support:
pip install langchain-togetherto use models from TogetherAIpip install langchain-anthropicto use models from Anthropicpip install langchain-fireworksto use models from Fireworks
We value and appreciate all contributions to MDCrow.