Recently, many research groups have been addressing data-driven approaches for (retro)synthetic reaction prediction and retrosynthetic analysis. Although the performances of the data-driven approach have progressed due to recent advances of machine learning and deep learning techniques, problems such as improving capability of reaction prediction and the black-box problem of neural networks persist for practical useby chemists. To spread data-driven approaches to chemists, we focused on two challenges: improvement of retrosynthetic reaction prediction and interpretability of the prediction. In this paper, we propose an interpretable prediction framework using Graph Convolutional Networks (GCN) for retrosynthetic reaction prediction and Integrated Gradients (IGs) for visualization of contributions to the prediction to address these challenges. As a result, from the viewpoint of balanced accuracies, our model showed better performances than the approach using Extended-Connectivity Fingerprint (ECFP). Furthermore, IGs based visualization of the GCN prediction successfully highlighted reaction-related atoms.
Prediction and Interpretable Visualization of Retrosynthetic Reactions using Graph Convolutional Networks
Posted by
Yasushi Okuno
on 12 09 2020
Related content
03 10 2022
< 1 minute
Calculate on the cloud
In order to increase the flexibility, access and integrability, Calculators and Predictors have...
12 05 2022
< 1 minute
Coupling stabilizers open KV1-type potassium channels
ABSTRACT: The opening and closing of voltage-gated ion channels are regulated by voltage sensors...
13 11 2021
< 1 minute
Cheminfo Stories 2021 Virtual UGM | Boost analytical experiments with phys-chem properties
TRY CHEMICALIZE Log in for videos & slides
13 11 2021
< 1 minute
Cheminfo Stories 2021 Virtual UGM | Enhancing Sibylla’s innovative drug discovery platform with ChemAxon
Sibylla's innovative pharmacological platform is aimed at discovering new drugs for untreatable...