Synthace’s ‘Design of Experiments’ Approach Enables Multi-Factorial Wet Lab Experiments
By Allison Proffitt
October 4, 2023 | INNOVATIVE PRACTICES AWARDS—Cultured meat may not be the standard use case for a Bio-IT World Innovative Practices Award, but the entry from Higher Steaks and Synthace caught the judges’ attention less for its application as for the innovation in experimental design. The approach was one of six winners in the 2023 Innovative Practices Awards announced in May at the Bio-IT World Conference & Expo.
Higher Steaks is a lab-grown meat company based in the United Kingdom on a mission to reduce traditional meat consumption and earn a 5% share of the global pork market by 2030. Their approach is to take cells from an animal source—a pig, in this case—establish stem cell lines, and proliferate and differentiate those cells in media in a bioreactor into recognizable and consumable pork products. It’s an ambitious goal and an expensive one. Higher Steaks needs an affordable, food-grade media optimized for certain cell lines. Commercial media has always been an expensive part of drug design—but it’s an even higher percentage of cost when the product is the price of bacon!
Higher Steaks calculated that there were 22 factors—and interaction profiles—they should optimize across for a better media solution. Alex Rimmer, a stem cell scientist at Higher Steaks and team-lead on the Innovative Practices Award-winning project, estimated that experiments to optimize media for the cells would necessitate “20,000 manual calculations”.
Rimmer chose a different approach: a design of experiments (DOE) approach that allows for multi-factorial experiment designs, looking at many factors in a single experiment that is set up in a matrix of multi-dimensional space. The company used JMP from SAS, a statistical design package that does high-dimensional experimental design, listing the conditions that should be met in each well. To actually conduct the experiments, Higher Steaks employed Synthace, a life sciences SaaS platform that specializes in digital experiments.
Synthace nominated the work for the Innovative Practices Award and—as Rimmer was not available to travel to Boston in May—Markus Gershater, Chief Scientific Officer at Synthace, presented the project to the Bio-IT World audience.
Design Space in Many Dimensions
Experimental design has traditionally entailed a very straightforward approach, explained Gershater. In order to find out whether a particular experimental factor is important, you hold all the rest of the variables constant and change that one factor. For example, temperature, time, and pH. But this approach assumes a 2D design space. “We’re making the assumption that the optimal temperature doesn’t depend on the pH or vice versa.”
The design of experiments (DOE) approach, instead, allows for multi-factorial experiment designs, looking at many factors in a single experiment that is set up in a matrix of multi-dimensional space.
“The thing about biology, is that they way that it has evolved… there are these interactions all over the place. It is an emergent system, which is just another way of saying that there are loads of different components that interact in an unexpected way whether it’s your brain, or your immune system, or a leaf on a tree—all of biology,” he explained.
Higher Steaks’ commercial media problem was a perfect use case for a DOE approach, and the JMP tool modeled how all of the experiments should be designed to accurately test the many options. Synthace facilitated the actual experimentation.
Synthace, “automatically works out all of the liquid handling steps, essentially. All of those transfers of every stock solution and the stock concentrations that are required, how much you’ll need of those stocks,” Gershater explained. “It converts all of those calculations and all of those liquid handling actions into automation scripts, so you don’t actually have to go in and do manual pipetting to make up all of these media.”
Gershater said that Synthace has been doing DOE experiments for a decade and the highest number of factors they’ve gotten to has been 27. With 22 right out of the gate, Rimmer’s first run was “not too shabby,” he quipped.
Synthace provided a preview of the prescribed 320 experiments in silico, “so the Higher Steaks can be confident in their desired outcomes before they get started in the lab,” the team wrote in their entry. Then, after running the first experiments, the Higher Steaks team could see which wells were growing more cells and iterate from there. “Subsequent designs will be improving the number of cells you’re generating progressively more,” Gershater explained.
Higher Steaks saw massive amounts of increased growth for 15x reduction in media costs, the company reported in their Innovative Practices Awards entry. The automation saved the company 6-9 months’ work, they wrote, and demonstrated the viability of DOE methodologies. Synthace’s automation.
These results are not “all the way they need to go,” Gershater acknowledged, but it’s an extremely valuable first step.
The results also prove a premise for which Gershater has been an advocate for years. “I’ve been arguing for many years—and have quite a lot of data to support this—that we’ve been doing biology wrong by doing one factor at a time.” Gershater—and the Bio-IT World judges—believe the Higher Steaks use case models a new, and needed, opportunity for life sciences research.