Engineered Opto-Active Polymeric Systems for Protease Detection

Overview

The detection of proteolytic activity in vitro is a promising field in understanding and identifying biological processes that occur within living cells. There is little research and experimentation on how to incorporate fluorophores within a polymer system that will allow for the detection of chemical reactions dynamically, hindering the understanding of biological processes in time-specific areas. Opto-active casts are formulated from the incorporation of DQ-gelatin (dyed-quenched gelatin) in a polyurethane (PUR) network. The DQ- gelatin will be conjugated into the network, allowing the polyurethane to act as a cast, ‘holding’ the protein in place for Trypsin (metabolic enzyme) to degrade the DQ-gelatin. When the fluorescein isothiocyanate (FITC) fluorophores in the DQ-gelatin are cleaved by Trypsin, the released fluorophores produce a measurable fluorescent emission that can be read via a plate reader. The issue lies in whether this system can be created to view cellular processes in real-time through the controlled environment of the polymer network. Fluorescent polymers, specifically polyurethane, are a promising platform for the real-time detection of protease activity due to their biocompatibility, controllable mechanical properties, and low reactivity with biological molecules. Through an ANOVA test, it was shown that there is a significant difference in the fractional difference between groups with and without Trypsin. This acts as a preliminary study to be expanded to other intrinsic fluorophores using proteases important in the regulation and detection of cancer, specifically MMP-9 activity, to be acted as a biosensor for biological cues in metabolic pathways. FASFA: A Solver for the Fuzzy Differential Equations Philip Naveen Mills E. Godwin High School, Richmond, VA This paper introduces fast adaptive stochastic function accelerators (FASFA) for gradient-based optimization of stochastic objective functions. It works using Nesterov-enhanced first and second moment estimates. The method is simple and effective during implementation because of its intuitive/familiar hyperparameterization. The training dynamics can be progressive or conservative depending on the decay rate sum. It works well with low learning rates and mini batch sizes. Experiments and statistics showed convincing evidence that FASFA could be an ideal candidate for optimizing stochastic objective functions, particularly those generated by multilayer perceptrons with convolution and dropout layers. In addition, the convergence properties and regret bound are some of the bests on online convex optimization framework. Future experiments could modify FASFA using bandit-based multi-start strategy.

Competition history

• JSHS 2023 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.