/tools

Madrigal is an open-source model designed to predict the clinical outcomes of drug combinations based on multimodal preclinical data. It provides tools for training and testing models, generating embeddings, and evaluating drug interactions, making it a valuable resource in the field of drug discovery.

The prtm repository provides a deep learning library for protein models, enabling tasks such as protein folding, inverse folding, and ligand docking. It aims to streamline workflows in protein design and structure prediction, making it a valuable tool in computational biology and drug discovery.

MolRL-MGPT is a code repository for a NeurIPS 2023 paper that presents a method for de novo drug design using multiple GPT agents in a reinforcement learning framework. It incorporates large molecular datasets and benchmarks for evaluating the generated molecules, making it a relevant tool in the field of molecular design and drug discovery.

Build Your Agent is an open-source initiative that provides a collection of intelligent agents designed for scientific research, focusing on materials property prediction and drug discovery. It aims to facilitate the development and deployment of AI-powered tools for various research workflows.

AA-Score is a tool that predicts protein-ligand binding affinity using an empirical scoring function based on various interaction components. It is utilized in virtual screening and lead optimization within the field of computer-aided drug discovery.

This repository provides tools for predicting binding affinity and generating kinase inhibitors using active site sequence representations. It includes scripts for training models and optimizing molecular structures based on predicted affinities.

OpenPharmaco is an open-source software designed for fully-automated protein-based pharmacophore modeling and high-throughput virtual screening. It utilizes deep learning techniques to enhance the pharmacophore modeling process, making it suitable for drug discovery applications.

BALM is a deep learning framework designed to predict binding affinities between proteins and ligands by fine-tuning pretrained language models. It utilizes the BindingDB dataset and proposes improved evaluation strategies for assessing model performance, making it a practical tool for early-stage drug discovery screening.

RGMolSA is a tool for ligand-based virtual screening that utilizes a new surface-based molecular shape descriptor derived from Riemannian geometry. It aims to predict potential new hits by comparing molecular shapes to those with known favorable properties, facilitating the drug discovery process.

PocketAnchor is a tool designed for learning structure-based pocket representations to predict protein-ligand interactions. It includes functionalities for predicting binding sites and affinities, making it relevant for molecular property prediction in drug discovery.

DiffDock-Pocket is a molecular docking model that utilizes diffusion processes to predict ligand poses in protein binding pockets while accommodating flexible side chains. It provides functionalities for training models, running inference, and visualizing docking results.

This repository contains source code for applying graph neural networks to improve molecular property prediction by leveraging both high-fidelity and low-fidelity data. It includes methods for transfer learning and provides access to multi-fidelity datasets for drug discovery and quantum mechanics.

RetroBridge is a Markov bridge model that facilitates retrosynthesis planning by predicting reactants for given product molecules. It utilizes a generative framework to learn distributions in a discrete state space, achieving state-of-the-art results in retrosynthesis tasks.

The PaccMann^RL repository provides a pipeline for predicting drug sensitivity and generating hit-like anticancer molecules using reinforcement learning. It includes tools for training multimodal predictors and generative models for both omic profiles and molecular structures.

PlaNet is a geometric deep learning tool designed to predict population responses to drugs by utilizing a clinical knowledge graph. It integrates disease biology, drug chemistry, and population characteristics to forecast drug efficacy and safety in clinical trials.

PyPLIF-HIPPOS is an advanced molecular interaction fingerprinting tool that enhances the analysis of docking results from AutoDock Vina and PLANTS. It generates customized interaction bitstrings from the 3D coordinates of ligands and proteins, facilitating molecular docking post-analysis.

HELM-GPT is a tool designed for the de novo generation of macrocyclic peptides using a generative pre-trained transformer model. It allows users to train models and generate new peptide structures, contributing to drug discovery efforts.

AutoMolDesigner is an open-source Python application that facilitates automated design and screening of drug-like molecules using a combination of chemical language models and automated machine learning. It includes modules for deep molecular generation and molecular property prediction, enabling researchers to efficiently conceptualize and evaluate new molecular candidates.

T-ALPHA is a deep learning model designed to predict the binding affinity of small molecules to protein targets. It utilizes a hierarchical transformer framework and includes features for uncertainty estimation, making it a valuable tool for drug discovery applications.

ChemInformant is a Python library that provides an all-in-one solution for retrieving chemical properties from the PubChem database. It supports batch processing and offers analysis-ready outputs, making it suitable for various applications in drug discovery and cheminformatics.

The RDKit MCP Server allows language models to interact with RDKit through natural language, providing agent-level access to its functions. It facilitates molecular property prediction and manipulation, making it a valuable tool for drug discovery and cheminformatics applications.

The Open Targets Tractability Pipeline assesses the tractability of potential drug targets based on Ensembl Gene IDs. It categorizes targets into various buckets for small molecules, antibodies, and PROTACs, providing valuable data for drug discovery efforts.

This repository serves as the source of truth for drugdesign.org, providing tools and resources for drug discovery, including molecular dynamics and modeling. It supports various aspects of cheminformatics and drug design, making it a valuable resource for researchers in the field.

DeepDrugDomain is a Python toolkit designed for the preprocessing and prediction of drug-target interactions and binding affinities using deep learning techniques. It offers extensive preprocessing capabilities and built-in benchmarks, making it a valuable resource for researchers in computational drug discovery.