Computed ECD spectral data for over 10,000 chiral organic small molecules.

Area of Science:

• Chemistry
• Computational Chemistry
• Data Science

Background:

• Determining the absolute configuration of chiral molecules is crucial for natural products, catalysis, and pharmaceuticals.
• Electronic Circular Dichroism (ECD) spectroscopy comparison of experimental and theoretical spectra is a reliable assignment method.
• Theoretical ECD spectrum generation using time-dependent density functional theory (TD-DFT) is a bottleneck.

Purpose of the Study:

• To address the lack of standardized and comprehensive ECD spectral datasets.
• To facilitate deep learning applications in ECD analysis.
• To accelerate the workflow for determining absolute configurations of chiral molecules.

Main Methods:

• Development of the Chiral Molecular Circular Dichroism Spectral (CMCDS) dataset.
• Systematic structural benchmarking of ECD data.
• Ensuring high standardization, scalability, and molecular diversity within the dataset.

Main Results:

• Introduction of the CMCDS dataset, consolidating fragmented ECD data.
• The dataset is designed for high standardization, scalability, and broad molecular diversity.
• CMCDS enables advanced deep learning applications for ECD analysis.

Conclusions:

• The CMCDS dataset significantly advances deep learning applications in ECD spectral analysis.
• It facilitates data-driven discovery and accelerates the determination of chiral molecule absolute configurations.
• This resource supports progress in natural products chemistry, asymmetric catalysis, and pharmaceutical development.