/tools

This repository is a curated list of papers that explore the use of deep learning techniques in protein design. It includes resources on benchmarks and datasets relevant to the field, making it a valuable tool for researchers in computational biology.

Chai-1 is a multi-modal foundation model for predicting the structures of various biomolecules, including proteins, small molecules, DNA, and RNA. It utilizes advanced techniques to achieve state-of-the-art performance in molecular structure prediction across multiple benchmarks.

DeePMD-kit is a deep learning package designed to create models for interatomic potential energy and perform molecular dynamics simulations. It interfaces with various deep learning frameworks and classical molecular dynamics packages, making it suitable for a wide range of molecular systems.

OpenMM is a high-performance toolkit for molecular simulation that utilizes GPU acceleration. It provides flexibility and efficiency for running molecular dynamics simulations, making it a valuable resource for researchers in computational chemistry.

OpenFermion is an open-source Python package that focuses on compiling and analyzing quantum algorithms for simulating electronic structures. It includes data structures and tools for manipulating representations of fermionic systems, making it a valuable resource for quantum chemistry applications.

AlphaGenome is an API that offers access to a model for predicting various functional outputs from DNA sequences, including gene expression and variant effects. It is designed for analyzing genomic data and provides tools for visualization and variant scoring.

This repository is an unofficial PyTorch implementation of AlphaFold2, a model that predicts protein structures from amino acid sequences. It utilizes advanced deep learning techniques to generate accurate structural predictions, contributing significantly to the field of molecular biology.

AlphaFold 3 - Pytorch is an implementation of the AlphaFold 3 model from Google DeepMind, designed to predict protein structures using deep learning techniques. It provides functionalities for training and evaluating models on molecular data, particularly focusing on protein structures.

TorchDrug is a PyTorch-based machine learning toolbox tailored for drug discovery, enabling users to predict molecular properties and work with graph-structured data. It provides a range of datasets and models for various tasks in molecular machine learning.

MDAnalysis is a Python library that facilitates the analysis of molecular dynamics simulations. It supports various simulation packages and provides algorithms for analyzing molecular interactions, making it a valuable tool for researchers in computational chemistry and molecular biology.

PySCF is a Python-based framework designed for quantum chemistry simulations, enabling users to perform various calculations related to molecular properties and behaviors. It supports density functional theory and other quantum mechanical methods, making it a valuable tool for researchers in computational chemistry.

Protenix is an open-source tool designed for high-accuracy biomolecular structure prediction, particularly for proteins. It includes related projects for protein design and benchmarking, enhancing its utility in computational biology and drug discovery.

DiffDock is a state-of-the-art molecular docking tool that utilizes diffusion models to predict the 3D structure of protein-ligand complexes. It provides a confidence score for its predictions and supports various input formats for proteins and ligands.

JAX MD is a library for accelerated and differentiable molecular dynamics simulations, allowing researchers to efficiently simulate molecular systems and compute energies and forces. It leverages JAX for hardware acceleration and automatic differentiation, making it suitable for advanced molecular simulations and experiments.

ProtTrans is a repository that offers pre-trained language models specifically designed for proteins, enabling tasks such as feature extraction, prediction, and protein sequence generation. It supports the bioinformatics community by providing tools for analyzing protein sequences and structures.

Open Babel is a versatile chemical toolbox designed for handling chemical data across multiple formats. It allows users to search, convert, analyze, and store molecular data, supporting a wide range of applications in chemistry and molecular modeling.

The Therapeutics Data Commons (TDC) is an open-source initiative that facilitates the development and evaluation of AI methods for drug discovery. It offers ready-to-use datasets, benchmarks for model comparison, and tools for predicting molecular properties and generating new biomedical entities.

MolecularNodes is an add-on for Blender that facilitates the import and visualization of structural biology data. It allows users to create animations from molecular structures and import molecular dynamics trajectories, enhancing the representation of proteins and other molecules.

This repository provides a collection of Jupyter notebooks designed to teach practical cheminformatics using open-source software. It covers various topics including molecular property prediction, generative molecular design, and machine learning models applicable to cheminformatics workflows.

Graphein is a protein and interactomic graph library that enables the creation of geometric representations of protein and RNA structures, as well as biological interaction networks. It supports various molecular types and provides functionalities for graph construction, visualization, and analysis, making it a valuable resource for molecular design and drug discovery.

Psi4 is an open-source quantum chemistry software package that performs efficient and high-accuracy simulations of molecular properties. It is designed for both users and developers, providing a flexible Python interface and supporting a wide range of molecular types.

DeepPurpose is a deep learning library that facilitates the prediction of drug-target interactions, drug properties, and protein functions. It supports various molecular encoding tasks and provides tools for drug repurposing and virtual screening.

The Materials Discovery: GNoME repository provides a dataset of over 520,000 novel stable materials and includes models for discovering new materials using graph networks. It aims to facilitate research in materials science by offering tools for exploring chemical systems and computing material properties.

SciPipe is a library for creating robust and flexible scientific workflows using the Go programming language. It is particularly suited for bioinformatics and cheminformatics applications, allowing users to design and execute pipelines that can process molecular data and integrate various command-line tools.