/tools

The aizynthmodels repository provides tools for training and evaluating models that predict synthetic routes for molecules. It is designed to assist in the synthesis prediction process, making it a valuable resource in molecular design and cheminformatics.

Rxitect is a de-novo drug design library that utilizes deep reinforcement learning and generative flow networks to create molecules with a focus on synthetic accessibility. It aims to improve the practicality of generated molecules by incorporating retrosynthesis-aware models into the design process.

This repository provides a program and datasets for fragment-based retrosynthesis planning, utilizing a curated dataset of chemical reactions. It aims to facilitate the design and generation of molecules through retrosynthetic analysis.

This tool provides supporting code for optimizing the retrosynthesis of gold nanoparticles using Bayesian optimization techniques. It includes a simulator for generating target distributions and iteratively improves the synthesis process through optimization methods.

ReAgentAI is an AI-powered chemical assistant that facilitates retrosynthesis planning and molecular visualization. It provides access to extensive chemical databases and supports various chemistry-related tasks, making it a valuable tool for researchers and chemists.

The Retrosynthesis-Reaction-Pathway is a web application that generates complete retrosynthetic pathways for given product molecules. It employs machine learning models to predict reactions and visualize molecular breakdown pathways.

The NeoChemSynthWave: Data project focuses on providing access to open datasets for computer-assisted chemical synthesis. It includes functionalities for downloading, extracting, and managing chemical data, which can aid in the design and analysis of molecular synthesis routes.

NNAASynth provides a framework for analyzing the synthetic accessibility of non-natural amino acids and their protected forms. It employs cheminformatics tools and deep learning models to generate amino acid derivatives, perform retrosynthetic analysis, and evaluate synthetic routes.

InterRetro is a code repository for a model that facilitates retrosynthesis planning using worst-path policy optimization in tree-structured Markov Decision Processes. It aims to improve the generation of synthetic pathways for molecules, making it a valuable tool in molecular design.

RetroChem is a machine learning-based tool designed for retrosynthesis, predicting potential reactants from given chemical structures. It features an interactive interface for molecule input and visualization, making it useful for chemists in planning synthetic routes.

The Organic Synthesis AI Assistant is a desktop application that helps organic chemists design and optimize synthesis routes for complex molecules using advanced AI techniques. It allows users to input target molecule structures and configure synthesis parameters to generate efficient synthesis plans.

RetroCHEM is a command-line tool designed for predicting single-step retrosynthetic reactions using graph partitioning techniques. It provides functionalities such as SMILES parsing and conversion to various molecular formats, making it useful for chemists involved in molecular design and synthesis planning.

This repository focuses on modifying the Aizynthfinder code to enhance its usability for end-users, specifically in the context of retrosynthetic analysis. It includes functionalities for loading stock data and utilizes trained models for generating synthetic pathways.

The AiZynthFinder Seq2Seq and Transformer Expansion Policies project enhances the AiZynthFinder tool by integrating SMILES-based retrosynthesis using advanced neural network architectures. This aims to improve the prediction of synthetic routes in drug discovery and chemical synthesis.

RetroDiff is a code and data repository for a paper on retrosynthesis using multi-stage distribution interpolation. It aims to facilitate the generation of molecular structures through advanced modeling techniques.

The Precedent Finder is a tool designed to analyze chemical reactions by utilizing the USPTO dataset. It allows users to input reaction SMILES and retrieve relevant precedents, facilitating the exploration of chemical reaction pathways.

Apollo is an open-source tool designed for biocatalytic retrosynthesis planning. It utilizes deep learning techniques and parallel Monte Carlo Tree Search to generate and optimize synthetic routes for molecules.

This tool assists in planning the retrosynthesis of target molecules by breaking them down into purchasable blocks based on template reactions and available precursors. It evaluates the synthesis process by providing metrics such as the number of steps and precursors required.

ACERetro is a computer-aided tool for planning chemoenzymatic synthesis using a synthetic potential score. It facilitates the design and generation of synthetic pathways for small molecules through retrosynthesis.

MultiAiZ is a tool that enhances AiZynthFinder for multi-target synthesis planning by identifying synthesis routes and reaction intermediates. It optimizes the synthesis process by ranking intermediates based on cost and utility, facilitating the design of convergent synthesis routes.

The NeoChemSynthWave: Chemistry project provides programming utilities for computer-assisted chemical synthesis, leveraging machine learning to enhance the synthesis process. It aims to facilitate collaboration among researchers from various backgrounds in the field of chemistry.

AIChemist is a frontend tool designed for retrosynthesis, utilizing AI to generate synthesis routes and conditions. It allows users to save reaction paths, indicating its utility in molecular design and chemistry workflows.

The RetroSynthesis API allows users to predict synthetic routes and reaction conditions for given target molecules. It includes functionalities for validating SMILES representations and checking the availability of chemicals, making it a useful tool for molecular design and retrosynthesis.

The shallow-tree repository provides a tool for predicting synthetic routes and scoring molecular designs, enhancing the efficiency of retrosynthetic analysis. It utilizes deep learning models to improve the speed and accuracy of predictions in drug discovery applications.