Water Solubility Predictions for 4,453 Organic Compounds Using QSPR Models

4,453 organic compounds have their water solubility modeled using quantitative structure–property relationships (QSPR). The dataset was created by A.P. Toropova using CORAL software, which applied the correlation intensity index and Las Vegas algorithm to improve model quality. The average coefficient of determination for the validation set is reported as 0.94 ± 0.01.

Use Cases

• Predicting environmental mobility of chemicals based on modeled water solubility.
• Assessing potential toxicity of organic compounds based on solubility relationships.
• Evaluating the effectiveness of the correlation intensity index and Las Vegas algorithm for QSPR model development.
• Benchmarking new solubility prediction methods against the published model set.

Strengths

• Contains data for 4,453 distinct organic compounds.
• Model validation achieved an average R² of 0.94 ± 0.01.
• Released under a permissive CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is small at 404.2 KB, indicating limited scope.

Provenance

Source

figshare

Collection Method

Modeled using quantitative structure–property relationships (QSPR) with CORAL software.

Time Range

null

Freshness

Last updated 2026-04-24 08:56:36; freshness should be verified.

Geography

null