Skip to content
#

solubility

Here are 41 public repositories matching this topic...

A machine learning pipeline for the prediction of aqueous solubility of small chemical compounds is proposed. Pairs of SMILES chemical structures and experimental values of the solubility are used to learn a model. A Flask application reads the SMILES chemical structure of a new compound and predicts the solubility missing value in logS units.

  • Updated Nov 28, 2020
  • Jupyter Notebook

Part of my work realized during the years as a PhD student. I was in the team MEMO of the CIRIMAT lab in Toulouse. I study via a multi-scale approach the influence influence of H, C, N impurities and point defects on the solubility and diffusion of the oxygen in the nickel (fcc phase). My works was under the supervision of Damien Connétable and …

  • Updated Dec 30, 2021

Benchmark of ten uncertainty-aware models across three molecular properties (ESOL, AqSolDB, lipophilicity), crossing model class with input representation under label noise, correlated noise and distribution shift.

  • Updated Nov 8, 2026
  • Python

Add this topic to your repo

To associate your repository with the solubility topic, visit your repo's landing page and select "manage topics."

Learn more