RetroTRAE
KNU LCBC (Kyungpook National University)
Single-step retrosynthesis prediction using fragment-based tokenization of atomic environments and a Transformer architecture. Mimics chemical reasoning by learning changes in atom environments between products and reactants.
Best For
Academic retrosynthesis research; fragment-level reaction understanding
License
Open Source (check repo)
Strengths
- +Fragment-based tokenization
- +Transformer architecture
- +Interpretable atom environment changes
Limitations
- −Single-step only
- −No route planning
- −Limited to USPTO training data
- −Academic prototype
R&D Pipeline Coverage
Related Tools
More in Retrosynthesis
AiZynthFinder
AstraZeneca Molecular AI
Multi-step retrosynthetic planning tool using Monte Carlo tree search guided by neural network policies. Recursively breaks down target molecules into purchasable precursors. Production-used at AstraZeneca.
TorchDrug
DeepGraphLearning (Mila / Université de Montréal)
PyTorch-based ML platform for drug discovery covering graph neural networks, geometric deep learning, knowledge graphs, generative models, and retrosynthesis. Provides unified API for property prediction, generation, and synthesis planning.
Stay updated on RetroTRAE
Weekly newsletter covering AI tool releases, benchmarks, and what practitioners actually use.