Exploring structure-property relationships in magnesium dissolution modulators

Abstract Small organic molecules that modulate the degradation behavior of Mg constitute benign and useful materials to modify service environment light metal for specific applications. The vast chemical space potentially effective compounds can be explored by machine learning-based quantitative structure-property relationship models, accelerating discovery potent dissolution modulators. Here, we demonstrate how unsupervised clustering a large number potential modulators structural similarities sketch-maps predict their experimental performance using kernel ridge regression model. We compare prediction accuracy this approach prior artificial neural networks study. confirm robustness our data-driven model blind modulating 10 untested compounds. Finally, workflow is presented facilitates automated chemicals with desired properties from commercial database. subsequently prove concept validation five chemicals.