Mathematical tools reconstruct ancient genetic code

The genetic code of the last universal common ancestor of all present life on earth, appears to be surprisingly advanced. In his PhD thesis, Peter van der Gulik provides mathematical tools to reconstruct this genetic code. He also proposes a new parameter to characterize the genetic code that provides information about the production of amino acids.

Publication date: 18-09-2019

The genetic code of the last universal common ancestor of all present life on earth, appears to be surprisingly advanced. In his PhD thesis, Peter van der Gulik (CWI) provides mathematical tools to reconstruct this genetic code. He also proposes a new parameter to characterize the genetic code that provides information about the production of amino acids. Today, Van der Gulik will publicly defend his thesis Considerations in Evolutionary Biochemistry at the University of Amsterdam.

When life on Earth was still in its very early infancy, its genetic make-up was already surprisingly advanced. According to mathematical models developed by CWI researcher Peter van der Gulik, the evolution of a key set of genetic molecules, called transfer RNA molecules, was already nearly complete around four billion years ago.

Calculating transfer RNA
Transfer RNA molecules play a crucial role in the production of proteins in all living cells. In his PhD thesis, Van der Gulik demonstrates that at least 23 types of transfer RNA might have been present in the last universal common ancestor of all life forms on Earth. With these molecules on board, says Van der Gulik, this ancestral life form basically contained all the major ingredients to transfer the kind of genetic information bits that are present in modern day organisms.

Theoretical organism
The last universal common ancestor (LUCA) that Van der Gulik used as a model organism, is a theoretical organism, proposed by scientists. LUCA is suggested to have lived four billion years ago. This single-cell organism gave rise to the current three domains of life: Archaea, Bacteria and Eukaryota, the latter of which includes animals and plants.

Error robustness
Van der Gulik also proposes a new method to research the error robustness of genetic ‘codons’. These codons are sequences of three DNA or RNA nucleotides that correspond with a specific amino acid or stop signal during protein synthesis. To keep errors in protein production to a minimum, the codons should contain as few errors as possible. That’s because a change in codon sequence can result in a cascade of errors down the production line of proteins, which might even lead to cancer and other serious diseases.

Calculating the shape of amino acids
But just how susceptible are codons to such errors? Currently, researchers determine this error robustness mainly by using a measure called polar requirement. This measure is based on the chemical interaction of amino acids and nucleic acids.

However, Van der Gulik argues that another way to measure error robustness is by calculating the shape of amino acids. Using this method, combined with polar requirement, yields a much better understanding of error robustness. For instance, in his PhD thesis, Van der Gulik shows that the middle position of a codon is a lot more error robust than researchers expect when they base their opinions on just polar requirement.

Computational methods
Computational methods have proven to be very powerful tools to study evolutionary biochemistry, says Van der Gulik. "My PhD research is the result of an intense collaboration between mathematics and computer science, as well as biochemistry and evolutionary biology. I am very happy that we are now able to understand genetic code even better than we did when I started my research."

Van der Gulik will publicly defend his thesis at the University of Amsterdam. His research was supervised by Prof. Harry Buhrman and Prof. Wouter Hoff, who acted as promoter, as well as Dr. Dave Speijer who acted as co-promoter. Van der Gulik performed his research at CWI's Algorithms and Complexity group.