Comparative analysis using machine learning methods for determination of pharmaceutical solubility in supercritical CO2 considering physical properties of drugs
Saad M. Alshahrani · 2026
Read Comparative analysis using machine learning methods for determination of pharmaceutical solubility in supercritical CO2 considering physical properties of drugs by Saad M. Alshahrani (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13581923.txt.
Description
Saad Alshahrani compares machine learning methods for predicting how well drugs dissolve in supercritical carbon dioxide, a fluid used in drug particle making. Standard testing can flatter a model, because rows for the same drug leak into both training and test sets. His 2026 study uses four inputs, molecular weight, melting point, temperature, and pressure, plus a stricter leave-one-compound-out check.
A gradient boosting model performed best overall. On the loose random split it scored near perfect, but its error roughly doubled when whole compounds were held out. That gap shows how row-wise splits inflate scores. The author concludes the model tracks solubility well within known data, yet warns that fresh validation and richer inputs are needed for truly new drugs.
Catalog record
| CREATOR | Saad M. Alshahrani |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |