English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In the last few years, we have seen the transformative impact of deeplearning in many applications, particularly in speech recognition and computervision. Inspired by Google's Inception-ResNet deep convolutional neural network(CNN) for image classification, we have developed "Chemception", a deep CNN forthe prediction of chemical properties, using just the images of 2D drawings ofmolecules. We develop Chemception without providing any additional explicitchemistry knowledge, such as basic concepts like periodicity, or advancedfeatures like molecular descriptors and fingerprints. We then show howChemception can serve as a general-purpose neural network architecture forpredicting toxicity, activity, and solvation properties when trained on amodest database of 600 to 40,000 compounds. When compared to multi-layerperceptron (MLP) deep neural networks trained with ECFP fingerprints,Chemception slightly outperforms in activity and solvation prediction andslightly underperforms in toxicity prediction. Having matched the performanceof expert-developed QSAR/QSPR deep learning models, our work demonstrates theplausibility of using deep neural networks to assist in computational chemistryresearch, where the feature engineering process is performed primarily by adeep learning algorithm.

Chemception: A Deep Neural Network with Minimal Chemistry Knowledge Matches the Performance of Expert-developed QSAR/QSPR Models