Quantifying how microbes grow is fundamental to areas such as genetics, bioengineering, and food safety. In a collaboration between the Schmid lab and Duke Statistics, Tonner and colleagues revisit the old problem of understanding microbial growth.

This problem, thought to be solved in the 1940s by Jacob and Monod, was actually far from understood. A new paper, published on October 26, 2020, in the journal PLOS Computational Biology, reports that random effects such as experimental variability, batch effects or differences in experimental material are all found to affect the estimation of the microbes’ growth parameters.

This interdisciplinary team was led by Peter Tonner, a CBB alum from the Schmid lab, who is currently a postdoc at the National Institute for Standards in Technology. Tonner and his collaborators developed a new model of population growth that enables more accurate estimates of the biological effect of interest, while better accounting for variation due to technical factors such as batch effects.

“I am proud of how this paper truly works across disciplines to advance the fields of both statistics and biology simultaneously,” said Amy Schmid, a professor in Biology and author on the paper.

This collaboration is far from finished: “We are working with a Masters in Data Science team through the information initiative Duke (iiD) to develop a graphical user interface for this model,” said Schmid. Very soon, we will not only be able to quantify microbial growth in much more accurate ways, but will also be able to easily visualize it.

By Marie Claire Chelini

CITATION: Tonner, P.D., Darnell, C. L., Bushell, F. M. L., Lund, P. A., Schmid, A. K., Schmidler, S. C. “A Bayesian non-parametric mixed-effects model of microbial growth curves”, PLOS Computational Biology, DOI: https://doi.org/10.1371/journal.pcbi.1008366.