Logo UHasselt



News & Calendar

Logo UHasselt Universiteit Hasselt - Knowledge in action


Those who collaborate will prevail    Jun 17, 2019

Those who collaborate will prevail
Jun 17, 2019


Prof. dr. Jori LIESENBORGS



Can we make the analysis of DNA-sequencing results faster, better and more efficient? Which processes can we automate? These questions sparked a successful collaboration between researchers prof. dr. Jori Liesenborgs from EDM and dr. Patrick Vandormael from BIOMED. The result? A new program that turns days of analysis into just a few minutes.       


Jori: “How can we help other sciences using computer science, that’s what computational science is all about. These days, computer technology is being used in almost every sector and discipline. However, we almost never use its full potential and therefore miss out on many opportunities like increasing the speed and optimization of processes or even automate them.

Patrick: “I am part of the research group of professor Veerle Somers at the Biomedical Research Institute (BIOMED). We study the role of auto-antibodies in auto-immune diseases such as rheumatoid arthritis and multiple sclerosis. We specifically look at two things: can we use auto-antibodies as a biomarker to detect these diseases in an earlier stage? And do they have an active role in the disease process? In order to find these auto-antibodies in the blood, we use a technique called phage display, which recently won the Nobel Prize for Chemistry.


Patrick: “Phage display is a method in which bacteriophages, small viruses that infect bacteria, are used to study protein-protein interactions. These bacteriophages can be marked by attaching human proteins to their surface, allowing researchers to answer questions like: Which human antibodies are going to bind to this protein? Which percentage of the patient population has antibodies against this specific protein? As phages consist solely of a DNA genome surrounded by a simple protein mantle, the protein being studied always carries its genetic material, which makes its identification using DNA sequencing quite straightforward. The problem is that the analysis of these specific DNA-sequences consists of many manual steps that can take up a large amount of time. This limited us to analyze about 4 sequences per hour. We were wondering if we could do this more efficiently.

Jori: “That’s where the Expertise Center for Digital Media (EDM) came into play. Our starting point was to learn about and understand the manual process and divide it into smaller steps. First, you get a DNA-sequence. Then, you need to detect the gene fusion between the phage and the human gene in order to determine where the part of the human gene starts. Subsequently, this human gene was screened against a large online database in order to identify it. After completing all these steps, they can then determine how the fusion protein was expressed on the surface of the phage. Once we fully understood this process, we were able to proceed and start writing the program.


Jori: “Actually, the question wasn’t as complex as we first thought. In just a few weeks’ time, we were able to write a first version using the input of Patrick. We then tested this pilot version to answer questions like: What went well? What could be further optimized? And for which problems did we still need to find a solution? … With these questions in mind, we kept adapting and fine-tuning the program until we were able to analyze a couple of hundred sequences in just a few minutes time.

Patrick: “The whole automation process went really quick and efficient. I was already convinced after the first version because, with every run, the translation to the final protein sequences was correct. I was really surprised though since I expected that our question was too complex and the process was not suitable for automation.


Patrick: “That’s definitely the time we gain by using this program. We literally went from 4 sequences per hour to hundreds of sequences in under 10 minutes with just the push of a button. That is a major leap forward. Now we can analyze more data and do a more in-depth analysis of additional aspects of the data that we find interesting. Moreover, the program is capable of doing a much more reliable analysis of the DNA sequences with a lower error margin in comparison to our manual analysis. Finally, thanks to the program, instead of analyzing only a couple of hundred sequences, we can now finally analyze the full output of our phage display screening, consisting of about 10 000 sequences, which we are currently planning to do. We are really grateful for that.

Jori: “It gives me great satisfaction to see how we helped and reinforced another scientific discipline using our knowledge of computer science. That’s what computational science is all about.