About

The E-Score project focuses on computing Global-regular and Global-end-gap-free alignment between any two protein sequences using their embedding vectors computed by state-of-art pre-trained models.

Instead of a fixed score between two pairs of amino acids(like BLOSUM matrices), the cosine similarity is calculated between the embedding vectors of two amino acids and used as the context-dependent score.

Thesis and Proposal

• Link to Thesis Paper
• Link to Proposal Paper

System Requirements

Recommended Python Version: 3.10
Recommended RAM: 24GB

• Each of the models needs about 8-12GB of RAM and as sequence length increases, RAM requirements do.

Installation

General:

# Virtual Environment Creation
python -m venv venv
source ./venv/bin/activate

# Install Requirements
pip install -r requirements.txt

SageMaker Notebook (Jupyter Lab):

• Run the notebook with !pip install -r requirements.txt in the Imports code block.

Models