Lipophilic Drug Solubility Predictions in Supercritical CO2 Using Ensemble Learning

A dataset supporting the prediction of lipophilic drug solubility in supercritical carbon dioxide (SC-CO2). It contains experimental and modeled solubility data for four drugs: Sirolimus, Tacrolimus, Rifampin, and Teriflunomide. The data was created by Arwa Sultan Alqahtani and last updated in May 2026.

Use Cases

• Training machine learning models for drug solubility prediction based on molecular descriptors and process conditions.
• Benchmarking new predictive models against the reported ensemble model achieving R² = 0.99.
• Supporting process design for supercritical CO2 systems in pharmaceutical manufacturing.
• Analyzing the relationship between thermodynamic features and solubility for lipophilic compounds.

Strengths

• Model performance is quantitatively reported with high accuracy (R² = 0.99, RMSE = 0.089).
• The dataset includes data for four chemically diverse, representative lipophilic drugs.
• The computational framework integrates molecular descriptors, thermodynamic features, and process conditions.

Limitations

• The dataset is very small (16.4 KB), indicating limited scope.
• Row count and specific column-level documentation are unknown; field semantics must be inferred after download.
• Data may reflect bias inherent to the specific four drugs and experimental conditions studied.

Provenance

Source

figshare

Collection Method

Likely contains experimental measurements and model predictions from the described computational study.

Freshness

Last updated 2026-05-13 05:53:41; freshness should be verified.