How Automation Can Give Bioengineering a Boost
How Automation Can Give Bioengineering a Boost
Complete the form below and we will email you a PDF version of "How Automation Can Give Bioengineering a Boost"
Genetic technologies have advanced to the point that digital and experimental approaches can work hand-in-hand. A new collaboration between Microsoft Research and London-based lab automation company Synthace has used a bioengineering approach called Design-Build-Test-Learn (DBTL) to create artificial genetic pathways using Synthace’s automation platform, Antha. We spoke to Emilie Fritsch, a field application scientist at Synthace, to find out more.
Molly Campbell (MC): Can you discuss the applications of developing artificial genetic pathways?
Emilie Fritsch (EF): The applications of developing artificial genetic pathways are manifold and diverse. Genetic engineering is widely used in the field of industrial biotechnology for the production of biofuel, chemical intermediates or biodegradable plastics. The development of new artificial genetic pathways has also been used in healthcare, with examples including the production of insulin, vaccines, hormones and antibodies. The application of artificial genetic networks is becoming more common in biosciences as it enables more intelligent control mechanisms for some of the aforementioned applications. This approach is providing industry a greater level of control, or in some cases, a better understanding of novel behaviors to the pathways they are engineering.
Ruairi Mackenzie (RM): What is the Design, Build, Test and Learn (DBTL) approach and how does it differ from other bioengineering approaches?
EF: The DBTL approach is an engineering principle which originated in more traditional engineering disciplines such as mechanical or electrical engineering and has more recently been adopted in the biosciences, notably in the field of synthetic biology. The development and optimization process relies on iterations of the DBTL cycle. It is well suited for automation and incorporates data modeling as part of the analysis to improve the characterization of biological systems, with the models used for predictive redesign in subsequent iterations of the DBTL cycle.
This approach tends to be more systematic and efficient than more classical bioengineering approaches, as it takes into account the complexity of biological systems.
RM: In this study, you used the DBTL approach to test gene circuits based on Quorum Sensing systems. Why are these systems of interest to research?
EF: In nature, many microorganisms use small signaling molecules called autoinducers to communicate with each other and determine their concentration. The process of producing and recognizing these signals is called quorum sensing (QS).
Understanding QS systems has a wide range of biotechnological applications, notably for the treatments of pathogenic biofilms or mixed-species fermentations.
QS systems are also good to study and model gene expression and regulation in complex biological networks and decipher design rules for cell organization during development.
RM: What were the main findings of the study?
EF: By combining Microsoft Research’s Station Bplatform and Synthace’s Antha, Microsoft Research was able to accelerate their DBTL cycle. This approach allowed them to refine new genetic circuits more quickly for the study of QS and apply machine learning and computational analysis of the results to generate improved predictive models.
RM: How did the use of automation through Antha change how the study was conducted?
EF: Antha enabled the flexible design, planning and physical execution of all the liquid handling steps. Antha generated all the instructions for the liquid handler that were used for the experiment and allowed us to easily change parameters dynamically without having to re-program the robot. This facilitated the rapid generation of new genetic networks and their characterization, providing substantial walk away time for the scientists at Microsoft Research. In addition, Antha’s digitally encoded protocols enabled data structuring for the subsequent inference and machine learning analyses.
RM: Why did liquid handling time increase using automation compared to manual pipetting? Would it not normally reduce this time?
EF: Several factors can influence the run time, but one is the larger deck space and therefore the larger distance that the pipetting head needs to cover to execute the liquid handling steps. Although the liquid handling time can increase, the automated process provides significant walkaway time for the scientists as well as increased robustness, reproducibility and traceability.
Emilie Fritsch was speaking to Molly Campbell and Ruairi J Mackenzie, Science Writer for Technology Networks