/tools

PXMeter is a toolkit for assessing the structural quality of biomolecular predictions, including proteins and small molecules. It provides multi-metric evaluations and supports both command line and Python API interfaces for efficient analysis.

CA RFdiffusion is a repository that provides training and inference code for a protein structure diffusion model. It generates protein backbones through a two-step process involving diffusion and refinement, making it a valuable tool for computational protein design.

Chemeleon is a text-guided diffusion model designed for generating crystal structures based on natural language descriptions or specified chemical compositions. It aids in material discovery by allowing users to explore and create crystal structures, making it a valuable tool in computational chemistry.

pepp'r is a package designed for the evaluation of predicted molecular poses, allowing users to compute various metrics to assess the quality of structure predictions. It supports a wide range of metrics applicable to small molecules, proteins, and nucleic acid complexes.

PAMNet is a universal framework designed for accurate and efficient geometric deep learning of molecular systems. It excels in predicting molecular properties, such as binding affinities and RNA 3D structures, and utilizes graph neural networks to enhance performance in these tasks.

GraDe_IF implements a graph denoising diffusion model for inverse protein folding, allowing for the generation of protein sequences based on structural information. It utilizes advanced machine learning techniques to optimize the design of proteins, making it a valuable tool in molecular biology.

HalluDesign is an all-atom framework that utilizes a structure prediction model to iteratively co-optimize and co-design protein sequences and structures. It allows for the design of new protein sequences based on structural hallucination, making it a valuable tool in molecular biology and protein engineering.

RNA3DB is a dataset of non-redundant RNA structures from the PDB, designed for training and benchmarking deep learning models focused on RNA structure prediction. It includes various RNA chains labeled with non-coding RNA families and provides tools for customizing and building the dataset.

DiffCSP-PP is an implementation for generating crystal structures constrained by space groups. It includes functionalities for training models on datasets and evaluating crystal structure predictions, making it a useful tool for molecular design and generation in materials science.

MegaFold is a system designed to accelerate protein structure prediction models, such as AlphaFold2 and AlphaFold3, by optimizing performance on heterogeneous hardware. It provides enhancements in training speed and memory efficiency, making it a valuable tool for researchers in molecular biology.

DMPfold is a deep learning tool for de novo protein structure prediction that uses iteratively predicted structural constraints to model protein structures from sequences. It is designed to extend the coverage of protein modeling across genomes and provides a framework for generating multiple structural models for a given protein sequence.

DMPfold2 is a fast and accurate method for predicting protein structures from sequence alignments. It utilizes deep learning techniques to generate models, enabling high-throughput exploration of uncharacterized proteins.

RNAPro is a state-of-the-art model for predicting the 3D structure of RNA molecules, developed in collaboration with experts from a Kaggle competition. It utilizes advanced techniques such as template modeling and multiple sequence alignment to enhance the accuracy of RNA structure predictions.

Porter5 is a tool designed for fast and accurate prediction of protein secondary structure in both 3 and 8 classes. It utilizes advanced machine learning techniques to enhance the prediction quality, making it a valuable resource for researchers in molecular biology.

PROTACFold is a toolkit that predicts and analyzes PROTAC-mediated ternary complexes using AlphaFold3 and Boltz-1. It provides methods for structure prediction, evaluation, and analysis, facilitating advancements in PROTAC drug discovery.

AlphaFold3-Conda-Install is a step-by-step guide for installing and configuring AlphaFold 3, a state-of-the-art tool for predicting protein structures. It facilitates the setup of the necessary environment and dependencies to run AlphaFold 3 effectively on compatible hardware.

AlphaFold-StepByStep provides a step-by-step guide to running AlphaFold2 and AlphaFold-Multimer for predicting protein structures. It includes processes for searching homologous sequences, generating unrelaxed models, relaxing them, and sorting by confidence scores.

The Triangle Multiplicative module is a PyTorch implementation that efficiently mixes rows or columns of a 2D feature map. It is utilized in AlphaFold2 for predicting protein structures, contributing to advancements in protein design and molecular modeling.

Rosetta-DL is a collection of deep-learning packages aimed at predicting and designing biomolecular structures, including proteins and antibodies. It includes tools for protein folding and sequence design, contributing to advancements in molecular biology and computational chemistry.

GCP-VQVAE is a model that converts protein tertiary structures into discrete tokens using vector-quantized variational autoencoders. It enables the generation and evaluation of protein structures, achieving state-of-the-art performance on various benchmarks.

This repository provides an implementation of AlphaFold 3 using PyTorch Lightning and Hydra, focusing on predicting protein structures from amino acid sequences. It includes functionalities for training and evaluating models that predict molecular structures, making it a valuable resource in the field of molecular biology.

The af_backprop repository contains modifications to AlphaFold that enable backpropagation through the model, facilitating advancements in protein design. It is associated with projects that enhance protein design accessibility and improve sequence alignment methods.

MSAGPT is a protein language model designed for predicting protein structures through multiple sequence alignment (MSA) generation. It achieves state-of-the-art performance in scenarios where natural MSAs are scarce, making it a valuable tool for protein design and analysis.

The 'alphafold_singularity' repository contains a Singularity recipe for running AlphaFold, a software that predicts protein structures. It facilitates the use of AlphaFold in high-performance computing environments, enabling researchers to perform protein folding simulations efficiently.