Prediction of coating thickness for polyelectrolyte multilayers via machine learning

Abstract

Layer-by-layer (LbL) deposition method of polyelectrolytes is a versatile way of developing functional nanoscale coatings. Even though the mechanisms of LbL film development are well-established, currently there are no predictive models that can link film components with their final properties. The current health crisis has shown the importance of accelerated development of biomedical solutions such as antiviral coatings, and the implementation of machine learning methodologies for coating development can enable achieving this. In this work, using literature data and newly generated experimental results, we first analyzed the relative impact of 23 coating parameters on the coating thickness. Next, a predictive model has been developed using aforementioned parameters and molecular descriptors of polymers from the DeepChem library. Model performance was limited because of insufficient number of data points in the training set, due to the scarce availability of data in the literature. Despite this limitation, we demonstrate, for the first time, utilization of machine learning for prediction of LbL coating properties. It can decrease the time necessary to obtain functional coating with desired properties, as well as decrease experimental costs and enable the fast first response to crisis situations (such as pandemics) where coatings can positively contribute. Besides coating thickness, which was selected as an output value in this study, machine learning approach can be potentially used to predict functional properties of multilayer coatings, e.g. biocompatibility, cell adhesive, antibacterial, antiviral or anti-inflammatory properties.

Utility of HumanIn VitroData in Risk Assessments of Influenza A Virus Using the Ferret Model

“…Of note, analyses repeated using rank correlations were found to have relative significance (RS- p ) and strength ( r ) largely similar to those obtained with the linear correlations (see Supplemental Methods). Calculation of predictive power scores (PPS) was employed as a complementary analysis approach to Pearson correlations, as this asymmetric nonlinear index can explore predictive nonlinear and asymmetric relationships between in vitro and in vivo data not possible by linear correlation analyses alone ( 26 , 27 ). PPS analysis of viral titer data in vitro generally agreed with the strength of correlations obtained with Pearson correlations (Table S4).…”

The Effect of Machine Learning Algorithms on the Prediction of Layer-by-Layer Coating Properties

“…The methodology for data collection, both from the literature and experimental data, was previously described in the paper by Gribova et al 33 Briefly, film thickness data from the literature were combined with the in-house data produced using a Quartz Crystal Microbalance with dissipation monitoring (QCM-D), and different features such as polymer concentration, molecular weight, charge density, etc., were included ( Table S1 ).…”

“…As an iterative process, where parametrization is easily achievable, LbL coatings provide an opportunity to transfer ML techniques to functional biomaterial design. As such, in the recent paper by Gribova et al, the authors used literature data and in-house generated experimental results to analyze the relative impact of 23 different coating parameters on the coating thickness . This is the first time the authors utilized ML in prediction of coating properties.…”

“…As such, in the recent paper by Gribova et al, the authors used literature data and in-house generated experimental results to analyze the relative impact of 23 different coating parameters on the coating thickness. 33 This is the first time the authors utilized ML in prediction of coating properties. However, in order to simplify the modeling process, Gribova et al assumed that coating thickness has a linear relationship with the number of bilayers.…”

A comparative evaluation of layer‐by‐layer assembly techniques for surface modification of microcarriers used in human mesenchymal stromal cell manufacturing

“…[26,27] Incorporation of heparin and collagen multilayers on cell culture surfaces has been shown to modulate soluble factors in hMSCs better than heparin or collagen alone. [28] Furthermore, since the layers deposited during LbL are on the nano scale, [29,30] the decrease in concentration in the polymer solution is negligible and the solution may be reused.…”