/tools

ProfileBFN is an implementation for steering protein family design through Profile Bayesian Flow. It allows for the generation and evaluation of protein sequences, making it a valuable tool in the field of molecular design.

This repository contains designs and evaluations for protein binders created during the AdaptyvBio competition. It utilizes various tools and models, including AlphaFold, to generate and assess the stability and binding potential of protein designs.

ISAMBARD is a tool designed for the analysis, model building, and rational design of proteins. It aims to facilitate the design of coiled-coil and collagen structures, making it relevant for molecular design applications.

ALLSites is a deep learning framework that predicts protein binding sites using a transformer-based architecture combined with convolutional encoders. It utilizes advanced optimization techniques and pre-computed protein embeddings to enhance the accuracy of binding site predictions.

FAPEloss is a Python implementation of the FAPE loss function used in the AlphaFold algorithm for predicting protein structures. It provides a framework for testing and optimizing the loss function, which is crucial for accurate protein design.

ProteinF3S is an implementation of a model that enhances enzyme function prediction by combining various protein data types. It provides processed datasets and pre-trained weights for inference, making it a useful tool in the field of protein design and bioinformatics.

This repository contains a Python script that generates helical wheel visualizations based on input peptide sequences. It is useful for visualizing the arrangement of amino acids in alpha-helices, aiding in protein design and analysis.

The cryo_fit tool is designed for fitting atomic models into cryo-EM density maps, which is crucial for understanding protein structures. It provides functionalities that aid in the analysis and visualization of molecular structures derived from cryo-electron microscopy data.

proteinviz is an open-source tool that predicts the 3D structure of proteins based on their amino acid sequences. It utilizes a protein folding model to generate PDB files and visualize the protein structures in a user-friendly interface.

TCRcube is a machine learning model designed for predicting the interactions between peptide-MHC complexes and T-cell receptors (TCRs). It utilizes pre-computed representations from ESM2 and AlphaFold2 to facilitate inference and training on molecular datasets, making it a valuable tool in the field of molecular biology.

TrDesign_partialhall is a tool that supports the design of protein structures with partial hallucination capabilities. It allows users to modify and generate protein sequences based on provided PDB files, facilitating the optimization and design of molecular structures.