SQuIRL: 133,883 Small Molecules with Quantum Chemical Properties

133,883 small molecules from the QM9 dataset, with properties calculated at the ωB97X-D/aug-cc-pVTZ level of theory. The dataset includes SMILES strings, molecular geometries, functional groups, vibrational spectra, and various electronic and thermochemical energies. It was authored by Anirudh Krishnadas and last updated on 2026-04 17.

Use Cases

• Predicting molecular dipole moments and polarizabilities based on electrostatic property data.
• Training models for infrared spectral analysis based on harmonic vibrational wavenumbers and intensities.
• Estimating thermochemical stability based on thermal corrections to energy, enthalpy, and Gibbs free energy.
• Classifying molecular symmetry or functional groups based on point group and functional group annotations.

Strengths

• Contains 133,883 molecules, providing a substantial sample size for model training.
• Includes a wide range of 23 calculated quantum chemical properties per molecule, such as geometries, spectra, and energies.
• Data is stored in a structured HDF5 format, which is efficient for hierarchical data access.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for specific modeling tasks.

Provenance

Source

Derived from the QM9 dataset (doi: 10.1038/sdata.2014.22).

Collection Method

Properties calculated at ωB97X-D/aug-cc-pVTZ level of theory from original QM9 structures.

Time Range

null

Freshness

Last updated 2026-04-17 09:25:37; freshness should be verified.

Geography

null