Predicting Adhesive Free Energies of Polymer-Surface Interactions with Machine Learning

Polymer-surface interactions are crucial to biological processes and industrial applications of polymers. A polymer’s composition significantly influences its structural and functional properties, such as its solubility and conformations. Here we propose a machine-learning method to connect a model polymer's composition with its adhesion to decorated surfaces. We simulate the adhesive free energies of 20000 unique coarse-grained 1D sequential polymers interacting with functionalized surfaces and build support vector regression models that demonstrate inexpensive and reliable prediction of the adhesive free energy as a function of the sequence. Our work highlights the promising integration of coarse-grained simulation with data-driven machine learning methods for the design of new functional polymers and represents an important step toward linking polymer composition with its derived properties.