What is it about?
Uses well known statistical methods to deduce how well a computational model of data can predict the properties of new molecules or materials
Featured Image
Why is it important?
There is much confusion over this relatively straightforward problem, and a plethora of papers have appeared proposing supposedly new and improved ways of validation a model. This has created unnecessary confusion in the literature, e specially for new scientists entering the field
Perspectives
I trust that this paper will be useful to many researchers in drug discovery and materials design who use statistical and machine learning methods to design or optimise molecules and materials. Unfortunately it has not yet stopped the publication of unhelpful papers that propose complicated methods for achieving this important task
Professor Dave A Winkler
La Trobe University
Read the Original
This page is a summary of: Beware of R 2 : Simple, Unambiguous Assessment of the Prediction Accuracy of QSAR and QSPR Models , Journal of Chemical Information and Computer Sciences, July 2015, American Chemical Society (ACS),
DOI: 10.1021/acs.jcim.5b00206.
You can read the full text:
Contributors
The following have contributed to this page
