/tools

ProteStAr is a tool designed to efficiently compress and archive collections of protein structure files, including formats like PDB and mmCIF. It offers high compression ratios and supports various file types, making it useful for managing large datasets in protein research.

MDAnalysisMolSSIWorkshop-Intermediate2Day provides materials for a workshop on using MDAnalysis, a Python library for analyzing molecular dynamics simulations. The repository includes tutorials on loading, manipulating, and analyzing molecular simulation data, making it a valuable resource for those working in computational chemistry.

MDAnalysisWorkshop-Intro0.5Day contains materials for a workshop that introduces participants to MDAnalysis, a software tool for analyzing molecular dynamics simulations. The workshop includes lectures and tutorials on system manipulation, distance calculations, and molecular nodes, aimed at enhancing understanding of molecular simulations.

This repository implements a reinforcement learning framework combined with active learning to optimize the selection of small-molecule drug candidates for in-silico screening. It enhances the efficiency of training by focusing on informative samples, thereby reducing computational costs.

ReLMole is a tool for molecular representation learning that utilizes two-level graph similarities. It includes datasets for training and fine-tuning models for predicting molecular properties and drug-drug interactions.

MolBokeh is a Python package designed for interactive visualization of molecular structures using Bokeh. It allows users to display molecular data alongside other metrics, enhancing the analysis and understanding of molecular properties.

ADMET_Prediction_Models provides tools for predicting various molecular properties related to drug absorption and toxicity. It includes functionalities for training models on SMILES data and predicting outcomes based on trained models, making it a valuable resource for computational chemistry and drug discovery.

SelfPAD is a framework designed to improve the prediction of antibody humanness by utilizing patent data. It includes functionalities for pre-training and fine-tuning models specifically for evaluating antibody sequences.

This tool implements parallel graph neural networks to analyze protein-ligand interactions, facilitating the prediction of binding affinities and other molecular properties. It includes data preparation scripts for both classification and regression models, making it suitable for molecular property prediction tasks.

The Twitter Retrosynthesis Bot is designed to perform retrosynthesis analysis for random molecules sourced from PubChem. It utilizes the AiZynthFinder tool to assist in generating synthetic routes, making it relevant for molecular design and cheminformatics.

REMD_analysis is a Python package designed for analyzing data from replica exchange molecular dynamics simulations. It includes tools for preparing simulation input files, visualizing state transitions, and performing free energy calculations using advanced sampling methods.

KIMMDY is a reactive molecular dynamics pipeline that integrates Kinetic Monte Carlo with molecular dynamics simulations to study bond scissions in proteins. It allows for the monitoring of bond elongation and calculates rupture rates to facilitate the simulation of dynamic molecular processes.

libgmxcpp is a C++ toolkit designed for reading and analyzing GROMACS simulation files, such as .xtc and .tpr. It facilitates the creation of trajectory objects for molecular dynamics analysis, including essential functions for periodic boundary condition calculations and distance measurements.

MolSysMT is an open-source library designed to work with molecular systems, offering a unified interface for converting, querying, modifying, and visualizing molecular data. It integrates with various molecular dynamics libraries and supports common structural operations, making it a valuable tool for researchers in computational chemistry and molecular biology.

Chemprop Factor is a tool designed for predicting molecular properties through matrix factorization and deep learning techniques. It aims to enhance the accuracy of property predictions in the field of computational chemistry.

The ADMET Prediction System utilizes deep learning to predict Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) properties from molecular structures. It employs Graph Neural Networks and provides mechanistic interpretations of the predictions, making it a valuable tool in preclinical drug development.

moo-denovo is a tool for automated de novo design of drug-like molecular libraries, leveraging deep learning and multi-objective optimization techniques. It allows users to optimize molecular descriptors and generate new molecular structures based on specified criteria.

This repository benchmarks active learning protocols for predicting ligand binding affinities using various datasets corresponding to different protein targets. It evaluates the performance of machine learning models in identifying top binders, providing valuable insights for computational drug discovery.

CuGBasis is a high-performance library designed for efficient computation of quantum chemistry density-based descriptors using CUDA and Python. It enables the calculation of molecular orbitals, electron density, and other quantum chemistry properties, making it a valuable tool for researchers in computational chemistry.

The spica-tools repository offers tools for setting up coarse-grained molecular dynamics simulations using the SPICA force field. It includes functionalities for mapping all-atom configurations to coarse-grained representations and generating input files for popular molecular dynamics software like LAMMPS and GROMACS.

This repository contains a reaxFF LAMMPS script designed to simulate the oxidation of nanomaterials, followed by thermal equilibration and tensile testing. It aims to investigate how oxidation impacts the tensile strength of materials, making it relevant for studies in molecular dynamics and materials science.

VASP-post-processing contains scripts for extracting and processing data from VASP simulation outputs. It allows users to obtain temperature data, atomic positions, forces, and calculate thermal conductivity, making it useful for analyzing molecular dynamics simulations.

The Bohrium Science Agent SDK allows developers to create AI-driven scientific agents that can perform tasks related to molecular dynamics simulations and bioinformatics analysis. It supports the orchestration of scientific workflows and provides a visual interface for molecular structure display.