Machine Learning Enables Optimal Design of Anti-Biofouling Polymer Brush Films

Credit: Tokyo Institute of Technology

A polymer brush film is composed of monomer chains grown in close proximity on a substrate. Monomers that look like ‘bristles’ at the nanoscale form highly functional and versatile coatings that can selectively adsorb or repel a wide variety of chemicals and biomolecules. For example, polymer brush films have been used as scaffolds for growing biological cells and as protective anti-biofouling coatings to repel unwanted organisms.

As an anti-biofouling coating, polymer brush It is designed mainly based on the interaction between monomers and water molecules. Although this simplifies the design, it has proven difficult to quantitatively predict the adsorption of biomolecules such as proteins to monomers. complex interaction involvement.

Now, in a recently published study ACS Biomaterial Science & EngineeringA research group led by Associate Professor Tomohiro Hayashi at the Tokyo Institute of Technology (Tokyo Tech) in Japan used machine learning to predict these interactions and identify the film properties that have the greatest impact on protein adsorption.

In their study, the team tested 51 different polymer To train the machine learning algorithm, we use brush films with different thicknesses and densities with five different monomers. We then tested several of these algorithms to see how well their predictions matched the measured protein adsorption. “We tested several supervised regression algorithms, including gradient boosting regression, support his vector regression, linear regression, and random forest regression, in order to select the most reliable and appropriate model in terms of predictive accuracy. says Dr. Hayashi.

Of these models, the Random Forest (RF) regression model provided the best agreement with the measured protein adsorption values. Researchers therefore used RF models to correlate the physical and chemical properties of polymer brushes with their ability to adsorb serum proteins and enable cell adhesion.

“Our analysis showed that the hydrophobicity index, or relative hydrophobicity, was the most important parameter. We then followed the thickness and density of the polymer brush film, the number of C–H bonds, and the net charge of the monomer. , and the density of the film was .. the monomer molecular weight On the other hand, the number of OH bonds was rated as less important,” emphasizes Dr. Hayashi.

Given the highly variable nature of polymeric brush films and the multiple factors that affect brushing, monomerprotein Employing machine learning as a method to optimize interaction, polymer brush film properties can provide a good starting point for efficient design of anti-biofouling materials and functional biomaterials.

A leap forward in biomaterial design using AI

For more information:
Debabrata Palai et al., Prediction of Serum Adsorption to Polymer Brush Films by Machine Learning, ACS Biomaterial Science & Engineering (2022). DOI: 10.1021/acsbiomaterials.2c00441

Quote: Machine Learning Enables Optimal Design of Anti-Biofouling Polymer Brush Films (3 August 2022) Retrieved 08/03/2022 from biofouling-polymer. html

This document is subject to copyright. No part may be reproduced without written permission, except in fair trade for personal research or research purposes. Content is provided for informational purposes only. Machine Learning Enables Optimal Design of Anti-Biofouling Polymer Brush Films

Back to top button