SParrOW takes flight
A one-two punch of in-house innovation and adoption for flexible spatial transcriptomics data analysis
Sometimes the best solutions are homegrown. SParrOW, a powerful tool for analyzing spatial transcriptomics data was initially developed by Lotte Pollaris in the team of Yvan Saeys at VIB’s Center for Inflammation Research. With support of the Spatial Catalyst, it has become a valuable resource for spatial data analysts both within and outside of VIB.
The journey of SParrOW began with Lotte Pollaris, PhD student in Yvan Saeys’ group in Ghent, who initially developed the tool to tackle messy spatial transcriptomics datasets. "I started cleaning up targeted spatial transcriptomics data from molecular cartography, but as I kept on receiving new data, I realized we needed a more robust solution," she explains.
What began as a side-project would grow into her PhD’s main focus: the development of a scalable tool for spatial data analysis. Her SPAtial transcriptomics Omics Workflow was built to enhance cell segmentation, leading to higher-quality data and more accurate single-cell annotations. It also offers visual quality checks, helping users to interpret their results and keep control over the data processing.
Of course Pollaris isn’t the only researcher confronted with the problem of spatial transcriptomic data processing. In this rapidly growing field, analyzing and making sense of increasingly large and complex data sets is a challenge/bottleneck for many researchers.
Enter the Spatial Catalyst team, who recognized SParrOW’s potential to serve the wider VIB research community.
“Our goal is to help VIB users and make spatial technologies accessible to them,” says Evelien Van Hamme, head of the Spatial Catalyst. “Naturally, we want to make sure they are using the best tools available for data processing.”
Win-win
It was Frank Vernaillen of the VIB Spatial Catalyst, who originally made the connection, remembers Pollaris: “Frank already collaborated with us and spread the word about what we were doing with SParrOW.”
Van Hamme: “While other groups have developed and published similar types of data processing pipelines, we’re convinced that the one Lotte and her colleagues at the lab of Yvan Saeys created is currently best in class—and the most future-proof tool.”
The Spatial Catalyst collaborated with Pollaris to scale up the capabilities of SParrOW, and expand its applicability and usability. Users can fine-tune key settings for their specific samples in an interactive way and efficiently compare different staining techniques and gene mapping strategies, making data analysis more precise and customizable. But the team also thought of more practical aspects, by making sure, for example, that the tool is easy to install.
“I am not a programmer,” admits Pollaris. “So it has been great to have these aspects taken over by experts. Once the Spatial Catalyst team got on board, we were able to make massive progress—in a way I would never have been able to do on my own.”
“It’s a win-win situation,” explains Van Hamme. “We can build on the fantastic work already happening within VIB, to make cutting-edge tools accessible to the research community. For Yvan Saeys’ group, the wider adoption of SParrOW is exciting, but as their core focus is on development, our team can help bridge the gap by supporting users and providing feedback to improve the tool further.”
Benjamin Rombaut, another PhD student in the lab, is now adding new functionalities to SParrOW, in close collaboration with Arne Defauw in Van Hamme’s team.
Hands-on
Key steps in SParrOW’s roll-out were the two-day hands-on workshops, recently organized by the Spatial Catalyst in both Leuven and Ghent to train dozens of researchers on how to use the tool. The training combined practical sessions on individual laptops with analyses conducted on the high-performance computing (HPC) system.
The organization itself was an ambitious undertaking,” explains Van Hamme, “getting everyone to simultaneously run the tools on their laptops and via the HPC was far from trivial, but thanks to the collaboration with Alexander Botzki and Janick Mathys of VIB’s Training and Conferences team, everything came together.”
The feedback from participants was very positive, and also for Pollaris, this large workshop was a highlight and proud moment. “Prior to the first training, we had a handful of users,” she recalls. “As we do more and more trainings and workshops, we have lots of new people using SParrOW. It’s an amazing feeling to see the tool you developed rolled out, making a difference in the research projects of others. And of course, user feedback is incredibly valuable—it helps us refine the tool and make it even better.”
Maintenance
The story of SParrOW’s success underscores the values of a strongly intertwined techno-scientific ecosystem like the one we work hard to cultivate within VIB. At the same time, it also highlights the challenges of maintaining computational tools.
Pollaris: “Developing tools like SParrOW does require a lot of resources, and it’s hard for a single lab to keep them up to date. PhD students are only in the lab for four or max five years before they move on to the next chapter, and they can’t really focus on tool maintenance itself, as that rarely yields publishable results.”
Similar to wet labs that ensure they keep their experimental protocols and know-how active as students and postdocs come and go, she says, dry labs need a way to keep their pipelines and analysis tools up and running.
Van Hamme echoes this sentiment, pointing to the growing demand for computational expertise at VIB: “We’re increasingly focusing on computational solutions, and the need for pipelines and tools like SParrOW is only going to grow.”
A sustainable strategy for maintaining and advancing these tools is essential—not only to support ongoing research but also to keep pace with the ever-growing complexity of spatial transcriptomics and computational biology as a whole.
Want to be kept up-to-date on our biotechnological news and stories? Join our community and subscribe to our bi-monthly newsletter.
Liesbeth Aerts
Read more
