/tools

The RXN reaction preprocessing repository provides tools for preprocessing datasets of chemical reactions, including standardization, filtering, and data augmentation. It facilitates the creation of flexible data pipelines for chemical reaction data, which is crucial for various molecular applications.

GAML is a Genetic Algorithm Machine Learning software package that automates the creation of force field parameters for molecular dynamics and Monte Carlo simulations. It emphasizes atomic charge development for solvent simulations and outputs files compatible with GROMACS, validating parameters against various physical properties.

Hotpot is a Python package designed to facilitate communication among various chemical and materials calculation tools. It integrates deep learning with traditional cheminformatics to predict molecular properties, generate new molecules, and optimize experimental conditions.

VespaG is a fast predictor of single amino acid variant effects using protein language models. It leverages a large dataset of variants to provide accurate fitness predictions, making it useful for protein design and analysis.

DTITR is a tool designed for end-to-end prediction of drug-target binding affinities using a Transformer architecture. It leverages self-attention mechanisms to analyze protein sequences and compound SMILES strings, making it a valuable resource for drug discovery.

QuantNBody is a Python package designed for quantum chemistry and physics, enabling users to manipulate many-body operators and wave functions. It facilitates the construction of various quantum models and operators, making it useful for theoretical research and educational purposes in quantum mechanics.

The dmrgscf repository provides an interface for using DMRG and DMRGSCF methods as solvers for CASSCF calculations in quantum chemistry. It allows users to perform advanced molecular simulations and analyze electronic structures of molecules.

This repository provides input files for the NequIP code, which employs graph neural networks to accurately predict interatomic potentials. It is designed for data-efficient modeling in molecular simulations, making it a useful resource for researchers in computational chemistry.

Chiron is a tool that implements differentiable Markov Chain Monte Carlo methods, enabling efficient sampling in molecular simulations. It is designed to facilitate the exploration of molecular conformations and properties through advanced computational techniques.

InSilicoQ is a quantum computation-based package designed for drug design and discovery, utilizing quantum algorithms for property prediction and molecule generation. It integrates machine learning techniques to enhance virtual screening and small molecule design.

libxdrfile is a library designed for reading xtc files generated by GROMACS simulations. It serves as a fork of MDAnalysis's implementation, providing improvements for easier integration with other libraries in molecular dynamics workflows.

UniSim is a unified simulator designed for time-coarsened dynamics of biomolecules, utilizing generative models to simulate molecular interactions. It provides tools for training and evaluating models on various datasets related to small molecules and proteins.

PRO-LDM is a framework that utilizes a conditional latent diffusion model to design and optimize protein sequences. It enables the generation of natural-like sequences with tailored properties, making it a valuable tool for protein engineering and molecular design.

DockCoV2 is a molecular docking pipeline that facilitates the prediction of binding affinities between FDA-approved drugs and SARS-CoV-2 proteins. It provides a comprehensive drug database and allows users to download docking data for further analysis.

This repository hosts tutorials for using GROMACS, a software package for molecular dynamics simulations. It includes various input files and step-by-step guides for simulating molecular systems, such as proteins and polymers, which are essential for understanding molecular behavior.

LSQT-Jupyter is a tutorial that demonstrates linear scaling quantum transport methods using Jupyter notebooks. It serves as an educational resource for understanding quantum transport in molecular systems, although it is not intended for high-performance computing applications.

The 'react-ocl' repository offers React components that integrate OpenChemLib for visualizing and editing molecular structures in formats like molfile and SMILES. This tool is useful for developers working on applications that require molecular visualization and manipulation.

The QED (Quantitative Estimation of Drug-likeness) module calculates a molecular descriptor that reflects the drug-likeness of compounds based on various molecular properties. It utilizes RDKit for cheminformatics and provides different weighting schemes for QED calculations.

This repository provides tools for exhaustive exploration of chemical space using a transformer model. It includes functionalities for generating molecular structures, preprocessing datasets, and computing molecular fingerprints, making it a valuable resource for molecular design and cheminformatics.

The PS4 Dataset is the largest open-source dataset for predicting protein single sequence secondary structure. It includes methods for validation and evaluation of secondary structure prediction models, making it a valuable resource for researchers in protein structure prediction.

The 'simple-md' repository provides code for running molecular dynamics simulations, aimed at simplifying the process. It includes instructions for building a container and running training scripts, indicating its utility in molecular modeling tasks.

LAGNet is a neural network designed for better prediction of electron density, specifically optimized for drug-like small molecules. It includes implementations of various models for this purpose, contributing to the field of molecular property prediction.