PHYSPROPNET: A Benchmarking Study of Machine Learning Models for Physicochemical Property Prediction in Data-Limited Environmental Research

Machine learning is increasingly applied in environmental chemistry for contaminant screening and property prediction, yet consistent benchmarks are lacking. We compared eight graph neural networks (GNNs) and nine conventional learning algorithms combined with five molecular descriptor and fingerpri...

Full description

Saved in:

Bibliographic Details
Main Author:	Joseph Wasswa (9253945) (author)
Other Authors:	Aryan Amit Barsainyan (20815671) (author)
Published:	2025
Subjects:	Genetics Biotechnology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Chemical Sciences not elsewhere classified Information Systems not elsewhere classified yet consistent benchmarks epa physprop database effective overall configuration applications requiring transparency richer molecular context moderate data sets larger data sets data set size data scale increases machine learning models environmental fate properties random forest matched physicochemical property prediction gnns achieved comparable five molecular descriptor property prediction molecular representation environmental chemistry descriptor models ∼ 1 scaffold splits rdkit descriptors outperformed gnns mordred features integrated accuracy increasingly applied higher accuracy high throughput contaminant screening composite ranking best suited benchmarking study 652 compounds 000 compounds
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

PHYSPROPNET: A Benchmarking Study of Machine Learning Models for Physicochemical Property Prediction in Data-Limited Environmental Research

Similar Items