Ku Leuven Winston Chiu氏インタビュー

October 17, 2022
Jasmin Lozza

“With the Caps-It we can rapidly perform large screens against highly infectious pathogens”

In this edition of Speaking with the Scientist we report on an inspiring conversation with Winston Chiu, a PhD student at the KU Leuven. He has published remarkable work on developing a high-throughput, high-content antiviral imaging assay for SARS-CoV-2 in a fully automated and isolated biosafety-level-3 (BSL-3) robotic system, the Caps-It (Chiu et al. 2022). Winston walked us through his experiences in setting up the Caps-It system and in using Genedata Screener® for the analysis of large data sets.

Winston, can you tell us about your current role at the KU Leuven?


I have a full-time job as the expert on our fully automated robotics system, called the Caps-It system. I am responsible for operational management, assay and instrument optimization and integration as well as implementation of high-throughput screening. In parallel, I am working on my PhD thesis in the Laboratory of Virology and Chemotherapy with special focus on viruses. My work addresses the question: how can we implement the Caps-It system for future pandemics? 

The rollout of COVID-19 vaccines was a milestone in the fight against the current pandemic. Why is the development of antiviral drugs still important? And how does your recent research contribute?

That is a question my family asks too. Luckily, we do have the vaccines. However, there is still a large group of people who cannot get vaccinated. Antiviral drugs might be a solution. Moreover, we know that the vaccine-induced immunity wanes and that the protection against certain variants is weaker. What happens if a new variant emerges and the vaccines do not work? In this case an antiviral drug can fill in. 

“Vaccines are very important but usually not enough to keep everything under control."

Our aim was to rapidly screen as many compounds as possible against SARS-CoV-2, to hopefully have an impact early in the pandemic. We screened over 2.5 million molecules in the last two years, and we could identify some hits, which were then subjected to further evaluation. At the beginning we screened many repurposing libraries provided by academic and pharmaceutical partners. One major collaborator was Johnson and Johnson, who created the VeroE6-eGFP cell line, which was massively used during the initial screenings.

These high-throughput screens were run on the Caps-It system. What exactly is the Caps-It? Can you briefly explain what advantages it provides, specifically during a pandemic?

The Caps-It-system is a fully automated robotic system inside an isolator. It works as a "human-out"- "pathogen-in" lab system and it is completely enclosed, which means that the operator does not enter the Caps-It and everything that leaves the system is first decontaminated. The dimension of the isolator is approximately 40-45 m3. The unique design allows researchers to perform experiments with live pathogens of high or unknown biosafety risk. The Caps-It is currently a BSL-3 laboratory, but at some point, we hope to be able to do BSL-4 research, or research on the most dangerous viruses. Also, every instrument is present in duplicate, so that if one instrument fails, screening can continue.

“With the Caps-It we can rapidly perform very large screens against highly infectious pathogens."

Watch this interesting video on the Caps-It system. 

What were the main challenges in putting together the Caps-it system?

The initial challenges came with designing and building the system, which were led by my advisor, Pieter Leyssen. Designing an isolated system for high-throughput screening on high biosafety level pathogens was not straightforward, because there are many things you have to take into account, e.g. defining decontamination procedures or processes in case of device failure. The system also had to be very flexible and adaptable because we cannot predict the next pandemic—we are sure that something else will come, but you don’t know what pathogen you are going to work with. Finally, the return on investment had to be maximal, and the system was designed for high-throughput screening as well as for the development of new and complex bioassays.

Now, for my part: the main challenge around implementing a new system was that there was no manual—just some technical drawings! I had to learn everything about the individual parts and instruments: the liquid handlers (a topic in and of itself), the plate readers and the high-content imagers. They all had to be harmonized and integrated. It took me almost two years to make the Caps-It fully operational, but it was ready by end of 2019—just before the onset of the COVID-19 pandemic. 

You screened thousands of compounds against SARS-CoV-2. Can you tell us more about the throughput of your system? 

We are a motivated team of young scientists and wanted to test the maximum throughput speed. The highest throughput we achieved was 164,000 compounds in 384 well plates in 10 days. However, this includes weekend work and evening shifts to manually move plates from and into the isolator, an effort we do not want to make every day.

In the recently published paper you comment that in developing high-throughput screening assays data storage, data acquisition speed, and logistics need to be considered. Can you elaborate? 

My view on this comes a bit from my background as an engineer. Once you go high-throughput and have hundreds of plates, data storage, acquisition speed and logistics require careful consideration, because you will sometimes have terabytes of data. You can say that your system can work 24/7, but how fast are you really acquiring and what are you going to do with such huge volumes of data? These are pragmatic questions that are much more important than just having the assay and screening. These questions are often overlooked by upper management. I like to compare it with my experience working in restaurants: what do you do if there is suddenly a table of thirty people, you need to have everything in place and ready, or at least know what you need to do to make it work.

Such a high throughput generates large amounts of data. What are the requirements for a software to process and analyze this data? 

I answer this question from a user’s point of view: Speed is very important—I have used programs before where you input data, but it takes a long time before you can do anything with it. Software needs to be practical and user-friendly.

You analyzed the data with the Genedata Screener software. What is the biggest advantage in using this software?

I really like the software and it was a huge advantage. It is fast. After an initial set up of the analysis template we can easily reuse it to quickly analyze new datasets. We also use Genedata Screener to get a quick overview of how the assay is going and for quality control—say, to see if something has gone wrong, like a bad reagent or bad batch of cells that we should toss. In this way the software is very helpful because with so much data it is simply impossible to check every plate individually. In addition, during the peak times of the pandemic, there was simply not enough time to do a proper quality control using standard software like Excel.

What is the future use of the Caps-It system?

We still do a lot of SARS-CoV-2 research because the pandemic is not yet over. However, the system can be used to do research on any pathogen. For example, we are currently working on a screen against the rabies virus, but we also plan to do research on bacterial infections, such as tuberculosis. Our future goal is to develop assays against pathogens that may have the potential to cause a pandemic and to gather as much knowledge and information as possible to be better prepared for the next pandemic.

What are your future plans?

I will stay in university for a little longer. I really like my research group, I am surrounded by geniuses and ultra-hardworking and motivated people, and that is truly inspiring. I see a huge potential in automation and in infectious diseases, two topics the world should focus more on and fields to which I would like to contribute.

Thank you very much for speaking with us, Winston.

Want to learn more about automation with Genedata Screener? Request the poster below:

Ask for Our Poster on End-to-End Automation with Roche  

The interview was conducted by Jasmin Lozza, Ph.D., Scientific Communication Specialist.
Pictures were provided by KU Leuven.