/tools

DeepChem provides an open-source toolchain that facilitates the application of deep learning in drug discovery, quantum chemistry, and biology. It supports various molecular tasks such as property prediction, molecular generation, and offers extensive tutorials for users to learn and apply these techniques.

NVIDIA Warp is a Python framework designed for accelerated simulation and spatial computing, enabling high-performance physics simulations and data generation. It supports differentiable programming, making it suitable for integration into machine learning pipelines, particularly in the context of molecular simulations.

Uni-Mol is a universal 3D molecular representation learning framework that supports various tasks such as molecular property prediction, binding pose prediction, and quantum chemical property prediction. It includes tools for molecular representation and docking, demonstrating state-of-the-art performance in these areas.

The 'tf-gnn-samples' repository provides TensorFlow implementations of various Graph Neural Network architectures. It includes tasks related to molecular property prediction, such as protein-protein interactions and quantum chemistry, making it a useful resource for researchers in computational chemistry and molecular biology.

SchNetPack is a toolbox that facilitates the development and application of deep neural networks for predicting quantum-chemical properties and potential energy surfaces of molecules and materials. It includes features for training models on benchmark datasets and supports molecular dynamics simulations, making it a comprehensive tool for atomistic machine learning.

TorchMD is an end-to-end molecular dynamics engine that leverages PyTorch to facilitate molecular simulations. It provides a user-friendly API for researchers to conduct force-field development and integrate neural network potentials into molecular dynamics workflows.

The BioNeMo Framework is a suite of tools and libraries optimized for training AI models in drug discovery, enabling efficient handling of biological data. It supports various molecular modeling tasks, including property prediction and molecular design, leveraging GPU resources for enhanced performance.

Uni-Fold is an open-source platform that advances protein modeling beyond AlphaFold, enabling accurate predictions of protein structures, including monomers and multimers. It provides tools for training and inference, making it a valuable resource for researchers in molecular biology and computational chemistry.

AtomWorks is an open-source platform that accelerates biomolecular modeling tasks by providing a toolkit for parsing, cleaning, and manipulating biological data. It includes advanced features for dataset featurization and sampling, making it suitable for deep learning applications in molecular biology.

LightDock is a docking framework that utilizes the Glowworm Swarm Optimization algorithm to facilitate the docking of proteins, peptides, and DNA. It allows users to define custom scoring functions and apply residue restraints, making it a versatile tool for studying molecular interactions.

BoFire is a Bayesian optimization framework designed for real experiments, particularly in the chemical and pharmaceutical industries. It supports optimization of molecular properties and experimental designs, utilizing advanced techniques like multi-objective optimization and active learning.

MolCLR is an implementation of a contrastive learning framework for molecular representation learning using graph neural networks. It enhances the performance of models on various molecular property prediction tasks and provides datasets for pre-training and fine-tuning.

gReLU is a Python library designed for training, interpreting, and applying deep learning models to DNA sequences. It provides functionalities for genome annotation and processing, making it a useful tool in the field of molecular biology.

Equiformer is a PyTorch implementation of an SE3/E3 equivariant attention network that achieves state-of-the-art results in protein folding. It utilizes advanced techniques in deep learning to model molecular interactions and predict protein structures.

IBM's FM4M is a multi-modal foundation model designed to support research in materials science and chemistry. It includes various pre-trained models for predicting molecular properties and generating molecular representations, making it a versatile tool for computational chemistry applications.

OpenComplex is an open-source platform designed for developing protein and RNA complex models, leveraging features from AlphaFold 2 and OpenFold. It allows for high-precision modeling and inference of RNA and protein-RNA complexes, making it a valuable tool in computational biology.

GAUCHE is a library that facilitates probabilistic modeling and optimization techniques tailored for molecular representations. It includes various kernels for molecules, chemical reactions, and proteins, enabling users to perform tasks such as property prediction and Bayesian optimization.

Matbench Discovery is an evaluation framework that ranks machine learning models on tasks related to high-throughput discovery of stable inorganic crystals. It predicts material properties such as thermodynamic stability and thermal conductivity, providing insights for building large-scale materials databases.

Bgflow is a PyTorch framework designed for Boltzmann Generators and other sampling methods, enabling accurate computation of free-energy differences and discovery of new molecular configurations. It integrates various sampling techniques, including molecular dynamics, to facilitate advanced molecular simulations.

The 'mdgrad' repository offers a PyTorch-based differentiable molecular dynamics simulator that allows for end-to-end simulations and optimization of molecular properties. It includes features such as automatic differentiation and compatibility with various force fields, making it suitable for advanced molecular simulations and research.

tblite is a light-weight framework designed for tight-binding quantum chemistry calculations. It allows users to perform single-point energy calculations and provides a high-level interface for manipulating parametrization data, making it useful for molecular property predictions.

PiFold is a tool designed for effective and efficient protein inverse folding, generating protein sequences that fold into specified structures. It employs novel features and a graph neural network approach to enhance the accuracy and speed of protein design.

Fermi.jl is a quantum chemistry program written in Julia that implements various post Hartree-Fock methods for energy computations and molecular property predictions. It is designed for high-performance calculations in computational quantum chemistry.

MolCRAFT is a series of projects aimed at developing deep learning models for structure-based drug design and molecule optimization. It introduces novel methodologies for generating molecules with high binding affinity and stable 3D conformations, addressing critical challenges in the field.