Leveraging the Power of Design of Experiments (DoE) For Optimising Growth Media Composition
Cell growth and proliferation are affected by a number of genetic and environmental factors and the interactions between them
All factors and their interactions need to be considered during the growth media optimisation process
The traditional one-factor-at-a-time (OFAT) method is reductionist and limits the number of factors that can be tested, resulting in some factors and interactions being ignored
Design of Experiments (DoE) is a powerful alternative that can test multiple factors and their interactions simultaneously in a single experimental run
By leveraging lab automation equipment and our software platform Antha, DoE enables scientists to explore a wider design space, accelerating the media optimisation process
When working with small organisms, such as cells, we tend to oversimplify the conditions needed for their growth and survival. However, just like humans, cells are impacted by multiple factors and the interactions of those factors. It is therefore crucial for scientists to be able to investigate all these conditions in order to optimise growth media during upstream development.
Traditionally, scientists have been using the one-factor-at-a-time (OFAT) approach , the only feasible approach for manual media optimisation. This method is reductionist as it studies each factor individually, ignoring interactions between them. In addition, with an increasing number of factors, it becomes impossible to study them all.
A more powerful alternative to OFAT is the Design of Experiments (DoE) approach that can investigate a number of factors simultaneously, taking into account their interactions. DoE can leverage lab automation and empowers scientists to explore a wider design space as computers can plan and calculate more sophisticated experiments than humans.
In a recent online demo, Scientific Consultant Luke Cach presented how our automation platform Antha can help biologists plan and execute sophisticated DoEs for media optimisation.
Oversimplifying Living Organisms Limits the Scope of Biology
Have you ever wondered why the boost of caffeine and sugar together seems stronger than each one separately? This outcome is known as synergistic interaction – the effect of the two products together is greater than the sum of each one individually. The complexity of human biology – our diet, environment, and social interactions – is a web of entangled factors, and our state of being is a result of their combinations.
Cells are also living things, and their reality is just as complex as ours. But when studying biology, we tend to simplify the matter of life: as the number of factors and their varying levels become overwhelming, we reduce them to just a few for simplicity.
Similarly, when developing media for cell growth and proliferation, we may choose to ignore the interactions between factors or just ignore a factor altogether. This is usually the case when we employ the reductionist OFAT method that only has the capacity for investigating one factor at a time.
Nevertheless, we need to account for all of the factors that can influence cell growth when developing media, in the same way we would have to account for all of the factors that affect humans. After all, we would not be able to model our ability to stay awake on just our caffeine or sugar intake alone – that is not how life works!
Utilising Design of Experiments for Growth Media Optimisation
Media optimisation exemplifies the need to uncover biology’s complexity without simplifying it. The source and concentration of media components may play a significant role in a cell’s ability to grow and proliferate. These media components should be considered not only independently but also in relation to other factors, whether genetic (e.g. promoters, repressors) or environmental (e.g. temperature, vessel shape, agitation rate).
With so many variables that could be tested at so many levels, the number of combinations can become daunting and is impossible to investigate using the traditional OFAT approach.
DoE is a technique that interrogates the effects of multiple independent factors and their interactions. It allows scientist to strategically test a select number of factor combinations and iterate upon the results to refine and optimise the design space in order to find the best conditions.
In combination with scale down models, DoE can provide a clearer strategy to screen and determine which factors and interactions have the greatest effect on cell growth. This approach can be iterated upon and carried through increasingly larger models, such as Ambr® 15 and Ambr® 250 bioreactors, for upstream development and, specifically, media optimisation.
Flexible Automation is Key to Efficient Design of Experiments
DoE is performed using lab automation, a combination of hardware, such as liquid handling robots, and underlying software. Unfortunately, automation comes with a range of challenges, such as the need for a solid programming background for operating the robots and the lack of flexibility in automation protocols.
Therefore, Synthace has developed Antha, a flexible and user-friendly automation software platform that overcomes these limitations and enables scientists to perform sophisticated DoEs with minimal training.
Using Antha, scientists can simplify the execution of their DoEs. They can provide a design file with the combinations of factors to be tested, while Antha figures out exactly how to physically test those combinations on selected automation hardware. It calculates the necessary volumes of stock solutions and how they should be mixed to make each unique combination, building individual liquid handling steps.
Antha also optimises DoE workflows for efficiency and consumable use and even provides an in silico preview of the execution steps and the deck layout of a selected robot. Scientists can thus validate their experiments in silico before Antha programs the robot for physical workflow execution.
Antha’s design is also iterative and flexible. The same DoE workflow can be applied to subsequent experiments by simply replacing the design file. Antha updates the workflow accordingly, allowing the user to test the newly desired design space. The scientists do not need to spend time on reprogramming the hardware or rebuilding the workflow.
See Antha at Work
DoE is a powerful approach to investigate biological complexity, but that does not mean it needs to be difficult. By planning, simulating, and optimising sophisticated DoEs and programming their execution on automation hardware, Antha reduces manual user interactions and the steps to create an automated process.
Leveraging the power of DoE, Antha enables scientists to optimise their growth media more effectively and allows them to use their brain power for planning meaningful experiments and analysing data instead of troubleshooting their experimental procedures.
To see Antha at work and learn how it enables sophisticated DoEs for media optimisation, watch the full demo. To find out how Antha was utilised by our clients for their DoE applications, check out our case studies with Oxford Biomedica, LabGenius, SPT Labtech. Or read our previous blog post to learn how Antha-automated DoE can be used for assay optimisation.
Machine Learning: It’s All About the Data
How to build a strong data foundation for machine learning applications.
Synthace Unveils First Life Sciences R&D Cloud Addressing Complexity, Speed & Reproducibility for Scientists
First ever no-code platform lowers barriers to automated biological experimentation and insight sharing.