Insights from Microscopy Experts

What We Learned from Meeting with Imaging Scientists & Imaging Software Fellows

A color-coded 3D projection of computationally separated cells from the eye of a developing zebrafish embryo as imaged using the lattice light-sheet microscopy with adaptive optics. Photo provided by CZI Imaging Scientist Srigokul Upadhyayula of UC Berkeley.

Microscopy underlies much of cell biology and biomedicine, and we believe that improving microscopy technology will accelerate efforts to cure, prevent or manage all disease. While the past decade has seen technical advances that have dramatically increased resolution of both light microscopy and electron microscopy, progress in the field has been slowed because of three major challenges arising from insufficient interaction between biologists and engineers.

Three challenges facing the microscopy field today

1. Many of the new microscopes, particularly when in the pre-commercial stage, require special expertise to assemble and even to operate.

We see value for biologists to focus on their expertise, engineers on theirs, and for the two to collaborate to solve problems that require cutting-edge technology. Biologists greatly value contact with both hardware and software engineers, but the biomedical infrastructure undervalues engineering. Engineers at imaging centers are often funded on a fee-for-service basis, which inhibits their creativity and underuses their expertise.

2. New imaging innovations spread slowly throughout the scientific community because commercialization is slow, or may not happen at all.

We see value in supporting optical engineers who can build pre-commercial microscopes and operate cutting-edge commercial instruments at U.S. imaging centers where they can actively teach the latest advances to scientists and students visiting from around the world.

3. The biological community struggles with visualizing, analyzing, storing, and sharing the large volumes of imaging data that modern microscopy can generate.

An existing ecosystem of tools meets some of the most immediate needs, but these tools remain complex to use and difficult to interoperate, and struggle with increasingly large data set sizes. We see several opportunities for CZI to support and improve this software ecosystem, and want more on-the-ground experience with these challenges to inform our thinking.

CZI’s two new programs were created to address these challenges head on. The Imaging Software Fellows Program directly supports developers of open source software for microscopy; the three Fellows are developers of scikit-image, FIJI / ImageJ, and CellProfiler.

The Imaging Scientists Program supports technical staff within 17 microscopy core facilities in the U.S. The expertise of these scientists ranges between hardware and software, but all share a passion for dissemination and teaching. Together, these two programs aim to create a community of experts who will interact with experimental biologists and with CZI to accelerate biomedicine through improvements in hardware and software used for microscopy.

CZI’s Imaging Software Fellows, from left: Allen Goodman of the Broad Institute of MIT and Harvard, Juan Nunez-Iglesias, of Monash University in Australia, and Curtis T. Rueden of the University of Wisconsin-Madison.

Learnings from our initial meeting

Grantees from the two imaging programs recently met in San Francisco — with each other and with other imaging experts from the CZ Biohub and Bay Area universities. The grantees will convene yearly to discuss advances and bottlenecks in the imaging field.

CZI Imaging Scientists and Imaging Software Fellows grantees at their first meeting together.

From this meeting, we learned about software needs of the microscopy field, the best ways to disseminate modern methods, and how to create community and foster learning. The following insights stood out:

1. The field could greatly benefit from a shadowing program.

By creating a system in which imaging experts can shadow other imaging experts, knowledge could be passed on more quickly through the field, and communication — both here in the U.S. and between the U.S. and other countries — would greatly increase. Because the Imaging Scientists manage cores and interact daily with experimental biologists, they are ideally suited to pass on their knowledge and teach non-experts. A shadowing program between the Imaging Scientists would increase their own expertise and create community.

CZI Imaging Scientist Michelle Itano participates in a job shadowing program.

2. Supporting robust computational tools is essential to move the field forward.

The use of light sheet microscopy is rising quickly, rapidly producing 3D images over time, with multiple colors and less photo damage. These very large, multidimensional data sets may take mere minutes to create, but weeks to analyze using present methods. Supporting computational tools will help solve this problem.

3. More support for open source software is needed.

Microscopy has many software needs, but the most profound concerns are the large data sets arising from multidimensional images: how to store, visualize, and analyze them. It is best to solve this problem with software that is freely available and open source to enable scientists to build on and reuse each others’ work. The Software Fellows Program supports individuals in this open source community, and members of the CZI computational biology and technology teams are identifying opportunities for contributing further. This meeting and others to follow will help keep the CZI team connected to the microscopy software community.

4. Microscopy courses that “train the trainers” would have the highest impact.

The Imaging Scientists (the trainers) are positioned in core facilities at universities to work with a wide variety of experimental biologists. They help decide the best optical devices for particular experiments and biological specimens, the many variables to optimize images, methods to record and store data, and software to visualize and analyze the data. The most effective way to disseminate all this complexity would be advanced courses where trainers get together to learn methods from each other, and subsequently share their knowledge with the users of their core facilities. Many of the grantees and other attendees at this meeting participate in such courses at the Marine Biology Lab at Woods Hole, Cold Spring Harbor Labs, and their own universities.

Where are we headed?

The Imaging Scientists and Imaging Software Fellows programs have multiple goals, and this meeting was a great start on one of the most important goals: to build community among a group of talented optical technologists. To follow up on the enthusiasm and momentum built at the meeting, we are polling participants about the design of a “shadowing” program (see learning #1 above) in which individuals might visit each other’s facilities for a week or so. In addition, we will encourage participation in the shadowing program at Euro-BioImaging in Heidelberg, Germany.

We are also thinking about how best to support our Imaging Scientists with the tools and resources they need to pass on their knowledge in the classroom. Teaching microscopy courses is an expectation of the Imaging Scientists and was a consideration in selecting them. Yet for us to be effective partners, we must first understand the courses in which our grantees are presently involved and identify if there are unmet needs.

How can CZI best support these existing courses? Are there courses that need to be created, either for specific subjects or for new communities?

CZI Imaging Scientist Caterina Strambio-De-Castillia of the University of Massachusetts Medical School writes on a blackboard.

To move the field of imaging forward, increased collaboration between biologists and technology experts is critical. With this in mind, we will continue to build community among our grantees and with other experts, aiming to better understand the challenges facing the field and how CZI can provide differentiated impact. Our support of the microscopy field is just beginning, and we’re excited to work with biologists and technology experts to accelerate science by improving microscopy tools.

Ed McCleskey, Science Officer

Ed McCleskey has been a Science Officer at the Chan Zuckerberg Initiative since March 2017, and he previously was a Science Officer for 10 years at the Howard Hughes Medical Institute. He was a professor and scientist at the Vollum Institute of OHSU (1993–2007) and at Washington University School of Medicine (1986–1993). His research addressed the biophysical properties of calcium-selective ion channels and the discovery of ion channels that trigger different types of pain. He taught physiology and neuroscience, and directed the neurobiology course at Woods Hole Marine Biology Labs.

Chan Zuckerberg Science Initiative

Written by

Supporting the science and technology that will make it possible to cure, prevent, or manage all diseases by the end of the century.