Ordinal Confidence Level Assignments for Regression Model Predictions

J Chem Inf Model. 2024 Dec 23;64(24):9299-9305. doi: 10.1021/acs.jcim.4c01755. Epub 2024 Dec 9.

Abstract

We present a simple method for assigning accurate confidence levels to molecular property predictions from regression models. These confidence levels are easy to interpret and useful for making decisions in drug discovery programs. We demonstrate their performance using time-split validation with assay data from the Relay Therapeutics internal database.

MeSH terms

  • Drug Discovery* / methods
  • Regression Analysis