Quantitative design rules for protein-resistant surface coatings using machine learning

Tu C Le,David A Winkler,Matthew Penna,Irene Yarovsky

doi:10.1038/s41598-018-36597-5

Abstract

Preventing biological contamination (biofouling) is key to successful development of novel surface and nanoparticle-based technologies in the manufacturing industry and biomedicine. Protein adsorption is a crucial mediator of the interactions at the bio – nano -materials interface but is not well understood. Although general, empirical rules have been developed to guide the design of protein-resistant surface coatings, they are still largely qualitative. Herein we demonstrate that this knowledge gap can be addressed by using machine learning approaches to extract quantitative relationships between the material surface chemistry and the protein adsorption characteristics. We illustrate how robust linear and non-linear models can be constructed to accurately predict the percentage of protein adsorbed onto these surfaces using lysozyme or fibrinogen as prototype common contaminants. Our computational models could recapitulate the adsorption of proteins on functionalised surfaces in a test set with an r2 of 0.82 and standard error of prediction of 13%. Using the same data set that enabled the development of the Whitesides rules, we discovered an extension to the original rules. We describe a workflow that can be applied to large, consistently obtained data sets covering a broad range of surface functional groups and protein types.

Highlights

The behaviour of proteins on surfaces is of critical importance in a wide range of applications, medical applications of nanomaterials[1], biomedical implants, artificial tissue scaffolds or industrial applications where surfaces are compromised when exposed to microbial or other biological contaminants[2,3,4,5,6,7,8]
A possible explanation for the relatively poor predictive power of these models is that the data set did not contain enough charged moieties to include them in the model, and the input parameters are all related to the ability of the functional ligands to interact with water and only reflect the hydration theory of protein repulsion[35]
Understanding the effect of surface chemistry on protein adsorption is critical for the design of novel bioinert materials

Summary

Introduction

The behaviour of proteins on surfaces is of critical importance in a wide range of applications, medical applications of nanomaterials[1], biomedical implants, artificial tissue scaffolds or industrial applications where surfaces are compromised when exposed to microbial or other biological contaminants[2,3,4,5,6,7,8]. Empirical rules that have been proposed to aid the design of protein repellent surfaces, the “Whitesides rules” are arguably the most widely used They arose from a systematic study of the protein adsorption capacity of 48 types of self-assembled monolayers (SAMs)[14,15]. This study aims to demonstrate the usefulness of statistical and machine learning techniques to identify quantitative relationships between the diverse chemistry of the material surface and the protein adsorption characteristics. These surfaces were functionalised with esters, ethers, amines, amides, sugars, nitriles and other functional groups. We use the same experimental data set from which the “Whitesides rules” were derived to illustrate that the technique can mine the data to extract established design rules quantitatively as well as initiate new rules

Objectives

Methods

Results

Conclusion