(left to right): Eliza Barkan, Madeleine Duran, and Amy Tresenrider
Editor’s note: This is the third in a series of Q&A articles on scientists engaged in the Seattle Hub for Synthetic Biology (Sea-Hub), a collaboration launched last December among BBI, the Allen Institute, and the Chan Zuckerberg Initiative. Here are observations from two Ph.D. candidates and a post-doc working in the lab of BBI’s Dr. Cole Trapnell, one of two scientific co-directors of Sea-Hub:
- Amy Tresenrider has been a Postdoctoral Fellow since 2019; she holds a Ph.D. in in Molecular and Cell Biology from the University of California, Berkeley.
- Eliza Barkan expects to complete a Ph.D. in Molecular & Cellular Biology later this year. She has worked in the lab since 2019 and holds a B.S. degree in Neuroscience with Minor in Chemistry from Bates College.
- Since 2020, Madeleine Duran has been a Ph.D. candidate and Graduate Research Assistant Advisor; she holds a Bachelor of Science in Computer Science and Molecular Biology from the Massachusetts Institute of Technology.
How have your education/experience prepared you for this project?
Amy Tresenrider (AT): My education overall has been pretty broad. But I have had an interest – from grad school and even undergrad – in gene regulation. I’ve been interested in how the chromatin landscape affects when and how genes are turned on. With Sea-Hub we aim to perturb many types of genes and pathways, but those that perturb chromatin formation and maintenance are definitely of interest. Additionally, while I learned how to analyze many different types of genomic datasets in graduate school, I also have a strong biological and experimental background which will be critical as we determine which perturbations to prioritize.
Eliza Barkan (EB): Before I started graduate school, I worked at the Allen Institute doing single cell transcriptomics to characterize the many different cell types in the brain. This experience made me realize how powerful computational biology can be, so when I started graduate school in 2018, I was drawn to the Trapnell lab. My Ph.D. work used a combination of experimental techniques in the lab and computational biology to identify signaling pathway dependencies in individual cell types and lineages during organogenesis – using zebrafish as a model.
Madeleine Duran (MD): I became interested in computational biology research during my junior year of college, so I decided to major in both biology and computer science. After graduation, I worked for two years at the Broad Institute in its Data Sciences Platform. Part of my job there was to write pipelines for running large-scale analyses for the Genomics Platform. After working in a very computational-focused environment, I was looking to switch to working closely with experimentalists. That is what drew me to the Trapnell lab. My Ph.D. work has been focused on writing software tools for analyzing atlas-scale single-cell perturbation data. My software, Hooke, provides a quantitative measurement of how cell types shift after perturbations (genetic, chemical, or environmental) as well as identifies which genes regulate those transitions.
Why did you want to work on this project?
AT: It’s a really great opportunity we have been given. The zebrafish is a pretty unique model. Compared to mice, it’s a vertebrate system that you can easily scale up in the number of embryos that can be collected and the number of perturbations that can feasibly be performed. They also develop very quickly, so there is a quick turnaround time from experiment to experiment. For these reasons, I think it’s the perfect organism to use for mapping out how developing vertebrates respond to varying genetic and chemical perturbations at the scale proposed by Sea-Hub. Hopefully, our work in zebrafish will integrate with the other arms of Sea-Hub to allow us to learn about the development of other species through our perturbations in the accessible zebrafish model. It’s just a very fun thing to be doing.
EB: Throughout my Ph.D., I have witnessed firsthand how the tools available to us largely dictate the types of questions that we can ask. This is why I know it is an exciting time for the Trapnell Lab – the experimental and computational tools the lab has recently developed, allow us to ask questions at a scale that we couldn’t previously, questions like “What are the dependencies of every cell type in every organ in a developing embryo on the many transcription factors and signaling pathways?” I am particularly looking forward to seeing my Ph.D. work that focused on using chemicals to inhibit entire signaling pathways, expanded, to identify when and in what combinations these signaling pathways are important as organs form. I can’t wait to hear about all the questions scientists around the world will be able to ask using the Sea-Hub public data resource.
MD: When I first joined the lab, we were in the early stages of doing our first sci-Plex zebrafish experiments. Throughout my Ph.D., it has been cool to see how that initial work has laid the groundwork for several new directions and projects in the lab. I am excited about the scale of data generation planned for Sea-Hub. It will allow us to be powered to develop and apply some new computational modeling tools.
Thus far, what has been surprising, unexpected, or challenging?
AT: For me the exciting thing is that Sea-Hub is a different type of science. As a post-doc, depending on one’s projects; you can become insulated. Whereas, with this project, the scope is so much wider. Collaboration and communication are essential. We want everyone to have the needed information to effectively complete the work that takes advantage of their strengths and specialties in a streamlined way which requires more investment in the logistics of figuring out how to work together.
EB: The single-cell transcriptomic data we are generating are large, so my role will be automating the analysis tools that Maddy Duran developed during her Ph.D. that analyze individual-embryo perturbation data. As we scale up the number of perturbations we are doing, I am establishing QC metrics and documentation, so scientists know they can trust both our developmental atlas and perturbation data. Then, they can then easily plug our data into their models to generate their own hypotheses about the biology that interests them. What I am actively thinking about now is finding the best ways to visualize and interactively publish our results, so scientists without a background in coding, can also interact with our data.
MD: A lot of our initial work at Sea-Hub is focused on streamlining our analysis workflow, so that we can focus more time on data interpretation. As someone who writes software, I believe it is useful to test my tools at this scale and receive feedback on their usability.
