Ferments concentration estimation
We work in conference with the department of Chemistry of Moscow State University. The main goal of the whole project is to create a soft/hard-ware complex to analyze patients blood samples, by estimating concentrations of specific enzymes. Classical Michaelis-Menten model is used.
Our group took a share in the project, by implementing various model fitting approaches, signal filtering methods. One of the most challenging problem here is to provide robust and fail-safe methods.
The first part is to deal with estimating enzyme-specific constants: Michaelis constant and rate constant.(The calibration).
The goal of the second part is to estimate concentration of each enzyme, given enzyme-specific constants.
In theory it is standard regression problem, however in our case standard approaches were unfeasible, due to noisy input data and ill separability of enzyme dynamics. Therefore we developed EM-like method for separating enzyme mixes from collection of measurements.
Have you spotted a typo?
Highlight it, click Ctrl+Enter and send us a message. Thank you for your help!