Motivation

A list of public datasets for QSPR or QSAR is crucial, as those types of datasets are quite rare and worthwhile. Herewith I list public datasets introduced in papers I’ve read recently. Hope these will help you develop a novel method.

List of public datasets

  1. Flash point MoleProp - github repository for the analysis
  2. Polymer Glass Transition Temperature in the supporting information
  3. Quantitative Structure–Activity Relationship Models for Ready Biodegradability of Chemicals
  4. Quantitative Prediction of Hemolytic Toxicity for Small Molecules and Their Potential Hemolytic Fragments by Machine Learning and Recursive Fragmentation Methods

Other hints

You may find other datasets in arXive or ChemRxiv, or either github repositories. Search them for those websites!