4 March 2019

Meet our experts: an interview with Jes Dreier, Image Analysis Specialist


Jes Dreier has recently joined DanStem/CPR as an Image Analysis specialist. During his career, he obtained a global understanding of microscopy – everything from building a microscope to having an image fully analyzed. In this interview, we asked Jes about the choices that brought him to DanStem and what are his ideas for the future of the Imaging platform.

By PhD student Carla Goncalves and Postdoc Ulf Tiemann.

People often overlook that research can be done in many different contexts, not just by group leaders or postdocs in academia. The way I see it is that you can do great research in a supportive role, where you enable others to push their research forwards and in that way become part of something bigger.

Jes Dreier, Image Analysis specialist

What first sparked your interest in biophysics?
The lab where I wrote my master thesis was located in a biophysics center and I was kind of a ‘one-man-team’, being the only one in the lab doing classical physics. As I decided to study for my PhD in the same institute, I wanted to be closer to the rest of the group’s research interests, so I started to look at things from a more biological point of view. The underlying physics was still the same but instead of looking at gold, I was suddenly looking at cell membranes. It was at this point that I realized using my knowledge of physics to describe the biological world was something that really captivated me.

Your field of interest during your PhD project was quite different from that of your postdoctoral work - what drove this switch?
During my PhD I had started to work with very advanced microscopes. We used polarized fluorescence microscopy, which resulted in images with a lot of information. A large part of my time was spent working out how to extract and precisely quantify different types of information from the imaging data set. Close to the end of my PhD I went to a conference where I met some of the people pioneering super-resolution microscopy. I knew this was the direction I wanted to go next. I had the great opportunity to help build the super resolution (STED) microscope in the lab of Jonathan R. Brewer at the University of Southern Denmark, in close collaboration with Christian Eggeling from University of Oxford.

What would you say is your greatest scientific achievement so far?
Live cell microscopy really appeals to me because if we can see individual protein dynamics inside the cell, we get very close to understanding how cells function. With this in mind, I joined Ilaria Testa’s lab in the Royal Institute of Technology, Stockholm, who has pioneered live cell imaging with super-resolution capability. During my time in Stockholm I worked to design and build the RESOLFT microscope, which uses reversible switchable fluorescence proteins to generate super-resolution in fluorescence microscopy. After about a year in the lab I had built every part of the RESOLFT and found living cells, or even worms, were ‘happy’ inside the machine  – meaning we could observe individual proteins in these cells in real time. It was so rewarding to have built this microscope myself and to see it perform so well in the task it was designed to do. The scientific ‘icing on the cake’ was to have this work was published in Nature Communications a month ago.  

You then took on a position as Image Analysis Specialist at DanStem/CPR, rather than keep on conducting your own research. Was this a logical next step for you?
I always want research to be part of my career and my current position definitely fulfills this need. People often overlook that research can be done in many different contexts, not just by group leaders or postdocs in academia. The way I see it is that you can do great research in a supportive role, where you enable others to push their research forwards and in that way become part of something bigger. After I finished my postdoc in Stockholm I knew that I didn’t want to go back to pure, “dead” physics. When I found the DanStem job posting, I realized that the role of image analyst was exactly what I wanted to do: bridging the gap between microscopy and biology.

You have been working at DanStem for four months now. How do you like it so far? What are the challenges?
Being here is really great and I enjoy the collaborations with people from different scientific backgrounds a lot. Since my previous expertise was more focused on microscopy than on pure image analysis, I had to familiarize myself with all the different complex software that people use here.  It’s been good to learn many new things but of course it takes some time.  Currently I think the greatest challenge is that people tend to come to me when a project is already well established and little can be done to change the experimental set up. I hope I can encourage researchers to ask for support when the question is still something like “should we install a fire alarm” and not “what should we do about the burning house”.

How would you define your role as DanStem’s Image Analysis Specialist? What are your ambitions here?
My goal is to enable people to do things they did not even know they could do. To achieve that, I have to get my knowledge out to the DanStem scientists. I want to teach them enough so they can do image analysis for their next project on their own in a much better way than they did before. And of course, they can always come back to me with more specific or complicated questions.

What are your plans to make this happen? What kind of training do you offer at DanStem?
I have recently organized basic and advanced courses and workshops for image analysis software programs such as Fiji and Imaris. But I also plan to launch a more informal coaching method. My idea is that people can drop by over lunch, or a piece of cake, to learn about a very specific technique, a small software tool, or a useful tip. I hope these ‘soundbites’ can spike interest and raise awareness about what is possible in image acquisition and analysis. If someone feels inspired by that to try something new, they can come back to me for more details and support.

From your perspective, what is the most exciting current development in the field of Image Analysis? How do you expect imaging-based research to change in the near future?
With the advent of machine learning and computer vision, many believe that these technologies will very quickly completely revolutionize the image analysis field. Actually, I do not share these high expectations because I see many limitations with the approach, such as the enormous amount of training data that is usually required. Nevertheless, I do expect machine-learning algorithms will become extremely useful in certain, specific applications. For example, scientists often have to manually identify specific structures, such as individual cells, in a digital image and label them for further computational analysis. Artificial neural networks can be trained to recognize and define such structures on their own, and hopefully this automated segmentation can be improved to work robustly even with weak signals and noisy data. If successful, scientists at DanStem would no longer have to sit in front of their computer screens for many hours, staring and clicking on thousands of images! Instead they can move to the fun part of analyzing their images with regards to a scientific question.

Learn more about our research platforms and facilities: