/tools

LLaMP is a multimodal retrieval-augmented generation framework designed to enhance large language models with high-fidelity materials knowledge. It allows for dynamic interaction with materials databases, making it useful for applications in materials informatics and molecular property prediction.

SumGNN is a tool designed for predicting multi-typed drug interactions by utilizing knowledge graph summarization techniques. It provides datasets and models for drug-drug interaction prediction, making it a valuable resource in the field of bioinformatics and drug discovery.

TypeDB Bio is an open-source biomedical knowledge graph designed to facilitate research in drug discovery, precision medicine, and drug repurposing. It allows researchers to query interconnected biomedical data, helping to identify potential drug targets and understand biological interactions.

StreaMD is a tool designed for high-throughput automated molecular dynamics simulations. It streamlines the process from system preparation to analysis, allowing users to run multiple simulations of various molecular systems efficiently.

Molecule-MCP is a server that connects various molecular science tools to Claude AI, enabling assisted molecule modeling and simulations. It allows users to control molecular modeling tools like PyMOL and ChimeraX through a unified protocol.

OpenFF Interchange is a project designed for storing and manipulating parametrized chemical systems, facilitating interoperability between different molecular simulation software. It provides APIs for exporting molecular data to common formats, supporting various force fields and chemical topologies.

This repository provides a codebase for designing molecules using structured state-space sequence models, enabling users to pre-train and fine-tune models for de novo drug design. It simplifies the process of generating new bioactive molecules with minimal code, making it a valuable tool for researchers in drug discovery.

DeepDFT is an implementation of a model that predicts charge density around molecular structures. It is trained on datasets containing quantum mechanical density files and can be used to predict electron density for new molecular configurations.

PharmacoNet is an open-source tool for protein-based pharmacophore modeling that utilizes deep learning for ultra-large-scale virtual screening. It automates the evaluation of ligands and supports feature extraction for deep learning applications in drug discovery.

ShEPhERD is a diffusion generative model designed for bioisosteric drug design, capable of generating new molecules in their 3D conformations based on learned distributions of molecular structures. It includes functionalities for training and inference, making it a valuable tool for molecular design and optimization.

ProteinDJ is a protein design pipeline that integrates multiple software packages to facilitate the design and optimization of protein structures. It allows users to generate new protein sequences and structures through various design modes, leveraging advanced modeling techniques like RFdiffusion and AlphaFold2.

The 'reaction_utils' repository offers utilities for handling datasets of chemical reactions, including template extraction and data manipulation. It is designed to facilitate the analysis and processing of chemical reaction data, making it a useful tool in the field of cheminformatics.

JuLIP.jl is a Julia library that facilitates the rapid implementation and testing of interatomic potentials and molecular simulation algorithms. It allows users to create and manipulate atomic systems, making it a valuable tool for molecular dynamics and materials science.

AlphaRED is a pipeline designed for predicting protein complex structures from sequences, utilizing AlphaFold for structure prediction and ReplicaDock for protein-protein docking. It provides a comprehensive approach to analyze and refine docking results based on predicted binding modes.

Calphy is a Python library designed for automated free energy calculations using LAMMPS as the molecular dynamics driver. It enables efficient computation of thermodynamic properties and phase transitions, making it a valuable tool for molecular simulations.

Grid is a Python library that facilitates numerical integration, interpolation, and differentiation, primarily aimed at supporting density-functional theory calculations in quantum chemistry. It is designed to assist researchers in the quantum chemistry community with various computational tasks related to molecular properties.

The ai2-kit is a command line and Python toolkit that integrates artificial intelligence with ab initio methods for complex chemistry systems research. It includes various tools for predicting molecular properties, conducting simulations, and developing automated workflows in computational chemistry.

Sparrow is a software package designed for fast semiempirical quantum chemical calculations, providing methods to calculate electronic energies, nuclear gradients, and transition dipoles. It is part of the SCINE framework, which focuses on chemical interaction networks and molecular simulations.

SECSE is a platform for systemic evolutionary chemical space exploration aimed at drug discovery. It utilizes deep learning and fragment-based design to generate novel small molecules, enhancing the hit-finding process in drug development.

Chemcoord is a Python module designed for transforming molecular structures between Cartesian and internal coordinates, specifically Z-matrices. It includes features for handling complex chemical structures and is optimized for performance, making it useful in theoretical chemistry workflows.

ChEMBL Downloader is a tool designed to reproducibly download, open, parse, and query the ChEMBL database. It simplifies the process of accessing a wealth of molecular data, which can be utilized for property prediction and other cheminformatics tasks.

xyz2graph is a Python package that converts XYZ molecular files into interactive 3D visualizations and NetworkX graphs. It allows users to analyze molecular structures and visualize them effectively.

gLM is a genomic language model designed to predict protein co-regulation and function. It provides tools for computing embeddings of protein sequences and allows users to train the model on custom datasets for various protein-related analyses.

ProGen2-finetuning is a tool for finetuning a generative protein language model to generate sequences from selected protein families. It allows users to preprocess data, finetune the model, and sample new protein sequences, making it relevant for protein design and generation tasks.