Interview | Shannan Ho Sui

Shannan Ho Sui is Director of the Harvard Chan Bioinformatics Core and a Principal Research Scientist in the Department of Biostatistics at the Harvard T.H. Chan School of Public Health. She is also an affiliated faculty member at the Harvard Stem Cell Institute, where she contributes to collaborative, cross-disciplinary research.

With a background in both biochemistry and computer science, Ho Sui found her path into bioinformatics during her PhD in Genetics at the University of British Columbia. Her research spans high-throughput sequencing, transcriptional gene regulation, and multi-omics data integration, with a particular focus on single-cell transcriptomics and epigenomics.

Passionate about fostering innovation through collaboration, she leads efforts to integrate computational methods into biomedical research, advancing areas like cancer genomics and immunotherapy.

At the Harvard Chan Bioinformatics Core, she is committed to promoting reproducible science, broadening access to cutting-edge tools, and supporting the development of strong interdisciplinary networks.
What motivated you to pursue a career in bioinformatics after completing your PhD in Genetics at the University of British Columbia?

I started during my undergraduate education, studying biochemistry with the goal of going into medical school. In my last year of my biochemistry degree, I started applying some bioinformatics methods to study archaebacteria, which surprised and intrigued me with the ability of using computational tools to help answer some of the key biological questions we were exploring. Back then, bioinformatics was just emerging as its own discipline.

I ended up pursuing a second degree in computer science because I was intrigued by these algorithms. While I was there, I met Dr. Fiona Brinkman, an infectious disease researcher working on pathogen genomics and listed in MIT’s Technology Review’s  “Top 100 innovators in the world under the age of 35”. She encouraged me to apply to do a PhD in bioinformatics in a brand new multi-disciplinary graduate program in bionformatics that the University of British Columbia and Simon Fraser University were jointly creating.

In your perspective, how can bioinformatics accelerate the discovery of new treatments for cancer?

I work at the School of Public Health, and I am also affiliated with Harvard Medical School. A lot of what we do is designed to address basic research questions that help us better understand a system or the development of a disease. We work on studies aimed at understanding genomic variation, changes in gene expression and gene regulation, and also try to identify potential drug targets—either to target immune suppressor genes or different cancer targets—but also to stratify patients based on their unique genetic backgrounds.

Bioinformatics allows us to take data of different types—such as genomic data, transcriptomic information (to examine expression patterns at both bulk and cell-type levels), proteomics data, and clinical data—and integrate them. This integration helps us gain insight into the mechanisms of cancer and also identify biomarkers, allowing earlier detection or the development of new treatments.

Related to that, there’s immunotherapy, which looks at how the immune system interacts with cancer. Bioinformatics plays a huge role in data sharing and curation—fostering collaboration across physicians and experimentalists—and enabling the development of new tools, including machine learning and AI, that leverage all this scientific information.

What are the main challenges you face as Director of the Harvard Chan Bioinformatics Core?

One of the main challenges—but also something I find very enjoyable—is interdisciplinary collaboration. As a core director, you have to be able to facilitate effective communication and collaboration between biologists, computational scientists, and clinicians. Even when working on similar problems, these professionals tend to “speak” slightly different scientific languages, which can be a barrier.

Another challenge is keeping up with rapid advancements, especially the new methods and technologies being developed. Within the Harvard Chan Bioinformatics Core, we’ve had to stay current because we want to democratize access to innovative methods for researchers. Keeping yourself and your team up-to-date is a critical component of running a bioinformatics core.

As a core, it’s also your responsibility to ensure that analyses are reproducible. So, there’s a strong emphasis on promoting practices that support reproducibility in bioinformatics research, as well as robust data management, to provide reliable information to our researchers.

In Portugal, research institutions like RISE integrate multiple fields without necessarily centralizing all researchers in a single large research center. Instead, they merge expertise from different institutions and disciplines. Do you see this decentralized, network-based model as an advantage or a challenge compared to the traditional large research center model? How can bioinformatics cores, such as those within universities, best adapt to and benefit from this structure?

I’ve worked a lot with the Harvard Stem Cell Institute, which is essentially a virtual institute—very similar to RISE—in that it is composed of multiple institutions and disciplines. I think there’s a lot of power in that. You get such diverse expertise, and having these different perspectives fosters innovation and enables you to address complex problems.

Another advantage is resource sharing. These decentralized structures allow each institution to benefit from collective resources, which helps establish strong collaborative networks and encourages the development of innovative ideas.

In cancer research, wet labs have traditionally received the majority of funding and recognition, while bioinformatics and computational biology are often underfunded and met with skepticism. This has led to fewer researchers specializing in bioinformatics, despite its growing importance. Do you think this funding imbalance and skepticism are related? How can we bridge the gap to ensure bioinformatics is better integrated and valued in translational cancer research?

Wet lab research often yields more immediately tangible results—visual data, physical samples, or direct clinical relevance—which can make it appear more “real” or trustworthy, especially to funders and researchers unfamiliar with computational methods. This perception can unintentionally marginalize bioinformatics, even though it’s essential for making sense of complex datasets, identifying patterns, and guiding experimental design. A significant part of the skepticism stems from a lack of understanding about the capabilities and rigor of computational biology. Many still view it as a supplementary tool rather than a driver of discovery in its own right.

To bridge this gap, we need to foster more integrated, cross-disciplinary collaborations where bioinformaticians are involved from the outset—not just brought in after the data is collected. Joint projects that highlight how computational insights can shape hypotheses, refine experimental focus, or uncover clinically actionable findings are critical. In addition, funding agencies should consider incentivizing these collaborations and supporting training programs that equip researchers with both computational and experimental literacy. This dual approach can help shift the culture and elevate the role of bioinformatics in translational cancer research.