Using Non-linear Regression to Predict Bioresponse in a Combinatorial Library of Biodegradable Polymers
- PDF / 141,110 Bytes
- 7 Pages / 612 x 792 pts (letter) Page_size
- 16 Downloads / 168 Views
JJ5.7.1
Using Non-linear Regression to Predict Bioresponse in a Combinatorial Library of Biodegradable Polymers Jack R. Smith1, Doyle Knight2, Joachim Kohn1, Khaled Rasheed3, Norbert Weber1, Sascha Abramson1 1
Department of Chemistry and Chemical Biology and the New Jersey Center for Biomaterials Rutgers, The State University of New Jersey New Brunswick, NJ 09803 2 Department of Mechanical and Aerospace Engineering and Center for Computational Design Rutgers, The State University of New Jersey New Brunswick, NJ 09803 3
Department of Computer Science The University of Georgia Athens, GA 30602 ABSTRACT We have developed an empirical method to model bioresponse to the surfaces of biodegradable polymers in a combinatorial library using Artificial Neural Networks (ANN) in conjunction with molecular modeling and machine learning methodology. We validated the procedure by modeling human fibrinogen adsorption to 22 structurally distinct polymers. Subsequently, the method was used to model the more complicated phenomena of rat lung fibroblast and normal human fetal foreskin fibroblast proliferation in the presence of 24 and 44 different polymers, respectively. In each case, the root mean square (rms) percent error of the prediction was substantially less than the experimental variation, showing that the models can distinguish high and low performing polymers based on structure/property information. Using this method to screen candidate materials in terms of specific bioresponse prior to extensive experimental testing will greatly facilitate materials development for biomedical applications. INTRODUCTION The past few years have shown rapid advance in the use of computational techniques to build, screen and mine libraries of compounds for molecular discovery and optimization.1,2 Yet, there has been little application to the design and optimization of biomaterials. This work represents the first such application to a combinatorial library of biodegradable materials. It has resulted in the development of an empirical method that can accurately predict the fibrinogen adsorption to and the rat lung fibroblast (FRLF) and normal human fetal foreskin fibroblast (NFF) proliferation on polymer surfaces. EXPERIMENTAL DETAILS Polymers from a combinatorial library of tyrosine-derived polyarylates were synthesized and characterized according to published procedures.3,4,5 In the fibrinogen adsorption
JJ5.7.2
experiments, films were solvent-cast into 384-well polypropylene plates and adsorption measured via immunoflourescence.6 In the FRLF study, the polymers were spin-coated onto glass cover slips that were inserted into the bottom of wells in 96-well plates. In the NFF study, polymers were solvent-cast into 96-well plates. The metabolic activity of FRLF and NFF cells (ATCC, Manassas, VA) was measured as described previously.3 COMPUTATIONAL METHODOLOGY Overview The modeling procedure was identical for each experimental data set. First, descriptors were generated for each polymer. Then, the significance of each descriptor with respect
Data Loading...
