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

Accumulation of molecular data obtained from quantum mechanics (QM) theoriessuch as density functional theory (DFTQM) make it possible for machine learning(ML) to accelerate the discovery of new molecules, drugs, and materials. Modelsthat combine QM with ML (QM-ML) have been very effective in delivering theprecision of QM at the high speed of ML. In this study, we show that byintegrating well-known signal processing (SP) techniques (i.e. short timeFourier transform, continuous wavelet analysis and Wigner-Ville distribution)in the QM-ML pipeline, we obtain a powerful machinery (QM-SP-ML) that can beused for representation, visualization and forward design of molecules. Moreprecisely, in this study, we show that the time-frequency-like representationof molecules encodes their structural, geometric, energetic, electronic andthermodynamic properties. This is demonstrated by using the new representationin the forward design loop as input to a deep convolutional neural networkstrained on DFTQM calculations, which outputs the properties of the molecules.Tested on the QM9 dataset (composed of 133,855 molecules and 19 properties),the new QM-SP-ML model is able to predict the properties of molecules with amean absolute error (MAE) below acceptable chemical accuracy (i.e. MAE < 1Kcal/mol for total energies and MAE < 0.1 ev for orbital energies).Furthermore, the new approach performs similarly or better compared to other MLstate-of-the-art techniques described in the literature. In all, in this study,we show that the new QM-SP-ML model represents a powerful technique formolecular forward design. All the codes and data generated and used in thisstudy are available as supporting materials athttps://github.com/TABeau/QM-SP-ML.

Molecular Design Using Signal Processing and Machine Learning: Time-Frequency-like Representation and Forward Design