Skip to content

Predict the whole sequence and 3D structure of masked protein sequences with ESM by @evolutionaryscale

License

Notifications You must be signed in to change notification settings

AstraBert/proteins-w-esm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Proteins with ESM

Predict the whole sequence and 3D structure of masked protein sequences with ESM by Evolutionary Scale.

How to use it

Locally - Build from scratch

The predictions are computationally-intense, you should have at least 16GB RAM and 4 core CPU

  1. Clone this repository:
git clone https://github.com/AstraBert/proteins-w-esm.git
cd proteins-w-esm
  1. Rename .env.example to .env and modify it with your Hugging Face access token (read or write permissions, fine-grained is nut supported)
  2. Launch the application and wait for it to load on http://localhost:7860
python3 app.py

Locally - Use Docker image

  1. Pull Docker image
docker pull astrabert/proteinswithesm:latest
  1. Run the image and wait for it to load on http://localhost:7860:
docker run -p 7860:7860 astrabert/proteinswithesm:latest

Online - Hugging Face demo

Use directly the Hugging Face Space demo that you can find here.

Online - Use GitHub CodeSpaces

Click the <> Code green button and then click on CodeSpaces -> Create CodeSpace on 'main'.

You should be redirected to a VSCode-based space connected to the port on which the app runs, directly accessible under the "Ports" section of the VSCode interface.