How Much Chemistry Does a Deep Neural Network Need to Know to Make Accurate Predictions?

By Garrett B. Goh, Charles Siegel, Abhinav Vishnu, Nathan Oken Hodas, Nathan Baker. 2017

48 citations recorded by the source corpus.

The meteoric rise of deep learning models in computer vision research, having achieved human-level accuracy in image recognition tasks is firm evidence of the impact of representation learning of deep neural networks. In the chemistry domain, recent advances have also led to the development of similar CNN models, such as Chemception, that is trained to predict chemical properties using images of molecular drawings. In this work, we investigate the effects of systematically removing and adding localized domain-specific information to the image channels of the training data. By augmenting images with only 3 additional basic information, and without introducing any architectural changes, we demonstrate that an augmented Chemception (AugChemception) outperforms the original model in the prediction of toxicity, activity, and solvation free energy. Then, by altering the information content in the images, and examining the resulting model's performance, we also identify two distinct learning patterns in predicting toxicity/activity as compared to solvation free energy. These patterns suggest that Chemception is learning about its tasks in the manner that is consistent with established knowledge. Thus, our work demonstrates that advanced chemical knowledge is not a pre-requisite for deep learning models to accurately predict complex chemical properties.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Charles Siegel
2. Abhinav Vishnu
3. Garrett B. Goh
4. Nathan Oken Hodas
5. Chaopeng Shen
6. A. Pais
7. Tânia F Cova

In Materials Chemistry

1. Charles Siegel
2. Abhinav Vishnu
3. Garrett B. Goh
4. Nathan Oken Hodas
5. Chaopeng Shen
6. A. Pais
7. Tânia F Cova

In Molecular Biology

1. Charles Siegel
2. Abhinav Vishnu
3. Garrett B. Goh
4. Nathan Oken Hodas
5. Chaopeng Shen
6. A. Pais
7. Tânia F Cova

Citations in the corpus

• Using Rule-Based Labels for Weak Supervised Learning: A ChemNet for Transferable Chemical Property Prediction (2017, methodology): “has shown that augmenting the image channels with basic atom and/or bond-specific chemical information improves overall performance [15].”
• SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Chemical Properties (2017, methodology): “To ensure that our results are comparable with contemporary DL models reported in the literature [42] and earlier work on Chemception CNN models [16, 17], we used the Tox21, HIV, and FreeSolv dataset from the MoleculeNet benchmark [42] for predicting toxicity, activity and solvation free energy…”