/tools/Rxitect
Rxitect
jcathalina/Rxitect
10 stars4 forksPythonAdded February 4, 2026
summary
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.
description
A collection of deep reinforcement learning-based & GFlowNet drug molecule generators focused on generation of molecules using Graphs/SELFIES guided by modern retrosynthesis tools to increase synthetic accessibility of de-novo designed drugs.
topics
drug-discoveryreinforcement-learninggenerative-modelmulti-objective-optimizationdeep-learningrnnselfiessmilesgflownetmolecular-graphstransformersynthesizability
Ratings
N/A
0 ratings
Rate this tool: