Routine and challenges of a microscopy unit

Hello Lucy. We are happy to meet you again; it has been five years since you started working with Argolight.

How did you become director of a microscopy team? What is the story of your career?

I have a Ph.D. in physiology, and I did a postdoc after that. I have always worked with microscopes; I have learned a lot about microscopy throughout my career. One day, I heard there was an open position here to manage a microscopy team. I applied, and now here I am. I started in another branch of the Institute before joining the cell science group. I manage a team of six people.

Are there recurring issues or subjects?

Yes, there are. We deal with two things on a daily basis: technical issues on the machines, and technical issues with the cells. We are trained biologists, we work in biotechnology, so we know that live things tend to have their own lives… and behavior!
We are a multidisciplinary team working together on the same project, so communication is always a big deal. We all have very different expertise. I work a lot with bioengineers who can overlap different areas: they can talk with the biologists and can very quickly switch and discuss code with the software engineers. This diversity in expertise and background is a real asset to the quality of our team’s output, but it also means we need to keep communication at the forefront and ensure we all understand the details of each other’s work.

What is the most common error that users have with your equipment?

All our users are internal; the instruments are not available to external users. Still, internal users may make some mistakes from time to time.
At the beginning, the first few errors were about the safety and upkeep of the device. We maintain 10 different microscopes on site, including Airyscan, spinning disk, Light sheet. We have a couple of microscopes that are kept on all the time and a couple that we do turn off. So sometimes, users may forget which is which. This was solved with user training.
Now, most of the errors are related to the software itself. When the microscopes’ manufacturers provide a workflow, our users can simply follow the steps, that is easy. But without a defined workflow, it is less consistent: there could be easy things forgotten, such as adjusting colors, or making sure you are using the right light path… We have had in the past some Zeiss and 3I technicians coming over to train us. We felt we had received sufficient training to train our own users, so we now do it ourselves onsite.

What is your team’s mission?

Our team is responsible for collecting data sets in a highly standardized manner. We have two main missions: building platforms and operating them. Our engineers build platforms, it is part of the experiment design, and after that, we produce data and do some preliminary processing on them for quality control. We are sometimes involved in the workflow until data analysis. Our data is packaged and handed over to more specialized teams, for instance, a modeling team that will use these large data sets. There are also assays that are being developed with the R&D team. We produce the data and pass them over to this team.
That is the bulk of what my team is doing. It is a unique role in some ways, as we are at the center of things.

What is your favorite part of the job today?

I like to solve problems when they happen. It is a big part of the job. I also enjoy designing new studies and new experiments. I have always liked that; it has not changed throughout the years. Maybe now, since I have more team members available to do the day-to-day work, I spend more time orchestrating the team.

How do you start a good day at work?

All of our days start with a scrum. Typically, people arrive at work, assess the situation, and after they have got this input, we run our scrum before launching the tasks of the day.
This evolved organically. At first, we were operating in start-up mode because we were developing all these different parts of the Imaging Pipeline. But that process has continued because there are a number of variables we cannot control. Every day, we need to assess what needs to be done. We see if there are issues that need to be addressed if we have healthy cells to work with, and so forth… Of course, we have weekly plans and long-term plans, but a daily approach is also necessary.

Thank you very much, Lucy.

Recommended reading:

• “Workflow and metrics for Image Quality Control in Large-Scale High-Content Screens”, M-A Bray, Anne Carpenter et al., Journal of Biomolecular Screening 17(2) 266–274, 2012 https://journals.sagepub.com/doi/pdf/10.1177/1087057111420292
• “Introducing a data-standard for fluorescence microscopy: increasing data quality and fidelity for biological measurements. Materials Science, Engineering.” M. Huisman, D. Grunwald et al., Proc. SPIE 10726, Nanoimaging and Nanospectroscopy VI, 107260A, Sept. 2018, https://doi.org/10.1117/12.2323660
• “An automated protocol for performance benchmarking a widefield fluorescence microscope.” M. Halter, J Elliott, et al., Cytometry A,;85(11):978-85, Nov 2014, https://doi.org/10.1002/cyto.a.22519
• Kurt’s Microscopy Blog, by  the former imaging from the director of the NIC@UCSF/QB3: http://nic.ucsf.edu/blog/