https://github.com/nvidia-bionemo-blueprints/generative-protein-binder-design
NVIDIA BioNeMo blueprint for generative protein binder design
https://github.com/nvidia-bionemo-blueprints/generative-protein-binder-design
biology drug-discovery nim nvidia
Last synced: about 1 month ago
JSON representation
NVIDIA BioNeMo blueprint for generative protein binder design
- Host: GitHub
- URL: https://github.com/nvidia-bionemo-blueprints/generative-protein-binder-design
- Owner: NVIDIA-BioNeMo-blueprints
- License: apache-2.0
- Created: 2025-01-09T20:58:51.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-03-05T17:41:14.000Z (3 months ago)
- Last Synced: 2025-03-23T18:44:50.275Z (about 2 months ago)
- Topics: biology, drug-discovery, nim, nvidia
- Language: Jupyter Notebook
- Homepage: https://build.nvidia.com/nvidia/protein-binder-design-for-drug-discovery
- Size: 322 KB
- Stars: 12
- Watchers: 3
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Security: SECURITY.md
Awesome Lists containing this project
README
# NVIDIA BioNeMo Blueprint: Protein Binder Design
The NVIDIA BioNeMo Blueprint for protein binder design shows how generative AI and accelerated NIM microservices can be used to design binders to a target protein sequence smarter and faster. This workflow simplifies the process of _in silico_ protein binder design by automatically generating binder sequences and predicted structures for the binder and target.
This Blueprint takes as input a valid amino acid protein sequence. It utilizes the following models:
- **AlphaFold2**: A deep learning model for predicting protein structure from amino acid sequence, originally developed by DeepMind.
- **ProteinMPNN**: a deep learning model for predicting amino acid sequences for protein backbones.
- **RFDiffusion**: a generative model of protein backbones for protein binder design.
- **AlphaFold2-Multimer**: A deep learning model for predicting protein structure of multimers from a list of amino acid sequences, originally developed by DeepMind.
Once completed, this Blueprint outputs predicted multimer structures (in PDB format) for the target protein sequence and any generated peptide binders. These binder-target multimeric structures can then be assessed to find binders that effectively bind the target protein.
## System Requirements
The docker compose setup for this NIM Agent Blueprint requires the following specifications:
- At least 1300 GB (1.3 TB) of fast NVMe SSD space
- A modern CPU with at least 24 CPU cores
- At least 64 GB of RAM
- Two or more NVIDIA L40s, A100, or H100 GPUs
## Get Started
- [Deploy with Docker Compose](deploy)
- [Deploy with Helm](protein-design-chart)
- [Source code](src)
## Quick Start
Deploy the blueprint using [Docker Compose](deploy) or [Helm](protein-design-chart)
```bash
cd ./src
jupyter notebook
```
## Set Up With Docker Compose
Navigate to the [deploy](deploy) directory to learn how to start up the NIMs.
## Set up with Helm
Follow the instructions in the [protein-design-chart](protein-design-chart) directory and deploy the Helm chart
## Notebook
An example of how to call each protein binder design step is located in [src/protein-binder-design.ipynb](src/protein-binder-design.ipynb)