Automated research accelerates scientific discoveries in yeast biology

Baker's yeast, Saccharomyces cerevisiae, is a well-studied organism that serves as, among other things, a model organism for cellular processes in human cells. A yeast cell contains thousands of different proteins, nucleic acid molecules, and metabolites that all interact in complex ways. However, there are still many knowledge gaps in yeast biology

"To accelerate scientific discoveries in yeast systems biology and gain a better understanding of how cells function, we need to dramatically increase the number of biological experiments in the field. Specifically, this means we need to find methods that allow us to characterize cells faster than before. In Ross King's research group, we are developing automated protocols for the selection of experiments and how to perform high-capacity analysis," says Daniel Brunnsåker, PhD student at the Department of Computer Science and Engineering.

Large numbers of parallel experiments

Daniel Brunnsåker is the lead author of a recently published study in which a group of researchers combined several methods, among them semi-automated experimental design, to characterize several genes involved in the process known as diauxic shift in baker's yeast.

The researchers also use metabolomics, a method used to study small molecules in organisms, i.e., the metabolites present in an organism. Information can be collected using various techniques, and in the current study, mass spectrometry was used.

Last but not least, the researchers use the robot scientist Eve, a laboratory system that uses AI to automate scientific experiments. Eve allows them to perform large numbers of parallel experiments with a high degree of automation. When combined with mass spectrometry, an analytical method with high automation potential, genes can be characterized in a rapid pace.

The diauxic shift − an extremely complex process

Daniel Brunnsåker explains that the yeast cells need to undergo a very significant change in their metabolism to succeed in the diauxic shift. Some microorganisms − and cancer cells in certain cases − can have multiple growth phases. In the case of yeast, cells typically grow on sugar while producing ethanol (alcoholic fermentation). When the supply of sugar is low, yeast cells slowly switch to consuming ethanol through respiration, and it is this shift that is studied by the researchers.

"It is an extremely complex process in which many different types of molecules are involved and influence the process. There may also be many genes that are important for the shift that we are still unaware of. Findings from our experiments could also help us understand the regulation of many other processes within cells, such as how cells protect themselves against oxygen radicals using antioxidants, as well as how organisms respond to different environments with varying conditions," says Daniel Brunnsåker.

The proposed method further characterized several genes with largely unknown functions, and more fully elucidated the role of several genes involved with the diauxic shift, several of which had human counterparts.

"Will be faster, and smarter"

The next step is to fully integrate the methods used in the study with the newest generation robot scientist, Genesis.

“We want to develop the method and make it faster, but also smarter. We are also working on integrating other types of biological information, not just metabolomics, to improve performance," says Daniel Brunnsåker.

Read more about the research

• Read the study in npj Systems Biology and Applications: High-throughput metabolomics for the design and validation of a diauxic shift model
• The study is partially supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP).
• Read more about Eve: Chalmers’ Robot Scientist ready for drug discovery