A new computer model can predict the speed of action of enzymes

Drug molecules and biofuels can be made to order by living cell factories, where biological enzymes do the work. Now researchers at Chalmers University of Technology have developed a computer model that can predict how fast enzymes work, helping to find the most efficient living factories, as well as studying difficult diseases.

The researchers tested their model by simulating the metabolism of more than 300 types of yeast. Against measured pre-existing knowledge, the researchers concluded that models with predicted kcat values ​​could accurately simulate metabolism. The image shows the common baker’s yeast, Saccharomyces cerevisiae. Image/Graphic/Illustration: Chalmers University of Technology

Enzymes are proteins found in all living cells. Their job is to act as catalysts that increase the rate of specific chemical reactions that take place in cells. Enzymes therefore play a crucial role in the functioning of life on earth and can be compared to nature’s little factories. They are also used in detergents and to make sweeteners, dyes and medicines, among other things. The potential uses are almost endless, but are hampered by the fact that enzymes are expensive and time-consuming to study.

“Studying every natural enzyme with laboratory experiments would be impossible, there are simply too many of them. But with our algorithm, we can predict which enzymes are the most promising just by looking at the sequence of amino acids that make them up,” says Eduard Kerkhoven, a systems biology researcher at Chalmers University of Technology and lead author of the study. ‘study.

Only the most promising enzymes should be tested

The enzyme turnover number or kcat value describes the speed and efficiency of an enzyme and is essential for understanding a cell’s metabolism. In the new study, the Chalmers researchers developed a computer model that can quickly calculate the kcat value. The only information needed is the order of the amino acids that make up the enzyme – something that is often widely available in open databases. Once the model has made a first selection, only the most promising enzymes should be tested in the laboratory.

Given the number of natural enzymes, the researchers believe that the new computational model may be of great significance.

“We see many possible biotechnology applications. For example, biofuels can be produced when enzymes break down biomass in a sustainable manufacturing process. The algorithm can also be used to study metabolic diseases, where mutations can lead to defects in the functioning of enzymes in the human body,” explains Eduard Kerkhoven.

More knowledge about enzyme production

The most possible applications are the more efficient production of products made from natural organisms, as opposed to industrial processes. An example is penicillin extracted from a mold, as well as the cancer-fighting taxol from yew and the sweetener stevia. They are usually produced in small quantities by natural organisms.

“The development and manufacture of new natural products can be greatly facilitated by knowing which enzymes can be used,” says Eduard Kerkhoven.

The calculation model can also indicate the changes in kcat value that occur if enzymes mutate and identify unwanted amino acids that can have a major impact on an enzyme’s efficiency. The model can also predict if the enzymes produce more than one “product”.

We can reveal if the enzymes have ‘moonlighting’ activities and produce unwanted metabolites. It is useful in industries where you often want to make a single pure product.

Eduard Kerkhoven, Systems Biologist, Chalmers University of Technology

The researchers tested their model using 3 million kcat values ​​to simulate the metabolism of over 300 types of yeast. They created computer models of how fast yeasts could grow or produce certain products, such as ethanol. Against measured pre-existing knowledge, the researchers concluded that models with predicted kcat values ​​could accurately simulate metabolism.


Chalmers University of Technology

Journal reference:

Li, F. et al. (2022) Deep learning-based kcat prediction enables improved enzyme-constrained model reconstruction. Natural catalysis. doi.org/10.1038/s41929-022-00798-z.

Maria D. Ervin