Optimization of Muscle Cell Culture Media Using Nonlinear Design of Experiments

Overview

Who: Zachary Cosenza, David E Block, Keith Baar

Published: August 13, 2021

Key Takeaway: Machine learning algorithms can reduce the number of experiments needed to solve bioengineering problems.

Research Topics: media

Summary

Zachary Cosenza et al. compare a hybrid nonlinear experimental design algorithm (HND) to a traditional design-of-experiments (DOE) method to test the functionality of machine learning algorithms in solving bioengineering problems. Their HND is used to formulate an optimal cell culture media for C2C12 cell proliferation and compared to the effectiveness of a similar cell culture media developed using a traditional DOE approach. Cosenza et al. find that HND is able to come up with a media comparable in efficiency to the media developed with DOE while requiring significantly fewer experiments. They argue this study is a valuable proof of concept for using machine learning in bioengineering optimization, as such algorithms can reduce R&D costs and timelines with further development.

Written by Morgan Ziegelski

View Open Access version

Go to Publication

Funding

Cite This Publication

Cosenza, Z., Block, D. E., & Baar, K. (2021). Optimization of muscle cell culture media using nonlinear design of experiments. Biotechnology Journal, 16(11), 2100228. https://doi.org/https://doi.org/10.1002/biot.202100228

Copied!