From learning to code to leading a BioHackathon team

Nobel Makonnen

Nobel Makonnen, a second-year student in the Applied Biomedical Data Science (ABDS) Master’s Program, who was placed in charge of a BioHackathon team this year.

“I signed you up to be a team lead for this year’s BioHackathon.”

Nobel Makonnen, a second-year student in the Applied Biomedical Data Science (ABDS) Master’s Program responded with stunned silence. Makonnen had intended to stay in the background during this year’s BioHackathon, a three-day event that brings together people across St. Jude to address pressing issues in biomedical data science for pediatric research.

His mentor, Stan Pounds, PhD, ABDS MS program associate dean and Department of Biostatistics member, believed that Makonnen was ready to be the program’s first student to lead a team at the marathon event. His statement took Makonnen a moment to process. “I was shocked when he told me,” Makonnen remembered. “Though I viewed it as my mentor having high hopes in my potential.” Then reality set in. “I felt so overwhelmed — how could I lead a team full of people more qualified than me?!”

Makonnen had been volunteered to oversee a group of seven scientists ranging from other master’s-level graduate students to postdoctoral fellows, all with more computational background than he had. A year earlier, Makonnen had participated in the event but was not quite ready to lead. This time, he drew on his graduate training to guide both a research project and the people working on it.

“The master’s program gave me not only the computational experience I needed but also the confidence in myself as a researcher,” Makonnen explained. “I had built the framework and foundation of the project for myself as a data scientist and could now talk to them as a peer about it.”

Building a computational foundation

When Makonnen applied to the ABDS program, he was nervous. While his undergraduate focus had been on chemistry and his research was in microbiology, he did not know how to code. To him, Python was a snake, and R was a letter, not the programming languages used in data science. However, St. Jude saw his potential and accepted him into the first inaugural cohort, despite the daunting transition.

“I was mortified of being thrown into something computation-heavy without knowing how to code,” Makonnen said. “In the beginning, it was bumpy, but I was given a lot of support, and by the end of my first year I felt very competent for where I was.”

Building that foundation is part of the curriculum’s design. First-year coursework develops computational skills through subjects such as biostatistics and machine learning. The remaining non-computational courses provide a background in other critical elements of a successful scientific career, such as ethics and public speaking.

In the second year, students join a faculty member’s lab and research a master’s thesis project. Makonnen chose Pounds’s lab, attracted by a biostatistics project aimed at helping researchers identify promising treatments for pediatric acute myeloid leukemia (AML).

A thesis project to find consistency in AML treatments

Childhood AML continues to be a challenging pediatric cancer to treat successfully. Motivated by that need, many scientists have attempted to find better treatments. The resulting body of literature is so large that comparing findings across that literature can be challenging.

“My project is to ultimately create a tool so researchers and clinicians can search for favorable AML treatments with confidence,” Makonnen explained. “I want to make a very speedy pipeline for them to find a list of potential therapies that performed well that they could then validate for themselves.”

The tool is being developed to help researchers identify promising AML therapies by systematically combining published preclinical and clinical research results to empirically predict the clinical outcomes of novel therapies, with an emphasis on identifying combinations with consistently favorable predictions. That comparison could provide researchers and clinicians a better place to begin investigating to find a more effective treatment approach. However, the activity of the new therapy would still need to be evaluated in future preclinical and clinical research.   

Nobel Makonnen and a team member

Makonnen talking with a team member about their work during the BioHackathon.

Makonnen brought his thesis project to the BioHackathon hoping to improve it through his teammates’ greater computational experience. However, before they could help develop the tool, everyone needed to understand the work he had already done. 

“I had intended to get them oriented to where I was, and move my project forward from there,” he said. “It took a long time to explain, and there was a lot of confusion, so nobody was on the same page … it was not going well.”

Leading from thesis to team project at the BioHackathon

Makonnen faced a choice: keep trying to bring everyone up to speed on his existing work or give the team a shared starting point. He chose to start again, with a smaller goal.

“Instead of trying to start from where I’m at, I changed our goal to rebuild everything from the very beginning so that it was not just my project; now it was everyone’s project,” Makonnen said. “Intentionally focusing on a smaller piece of the project established a more manageable and shared objective for the team, which went much better.”

Narrowing the scope gave the team a piece of the larger project to build together. Throughout the rest of the event, Makonnen moved among team members, asking and answering questions. His training had given him the technical grounding to participate in those conversations, even when his colleagues brought more coding experience. Beforehand, he had feared that he was unprepared to manage such a group, but as evidenced by his choice to pivot, he had gained the confidence and skills he needed to lead them in this computation-heavy environment from the ABDS program.

As the best leaders do, he also took the time to learn from these capable colleagues. One team member, Colleen Reilly, Department of Computational Biology, recommended a more application-like user interface, making the tool easier to use. She drew on her experience doing the same while helping develop ProteinPaint, a St. Jude data visualization portal deemed high-quality enough to be adopted by the National Cancer Institute’s Genomic Data Commons.

Others suggested ways to make his code more efficient and avoid technical problems as the project developed. Their contributions continued to shape Makonnen’s work long after the three-day event ended.

“I’ve been reviewing my old code to improve it,” Makonnen said. “I am slowly making progress, but as many of these fixes will make it easier to work further down the pipeline, they will have saved me time and made the end product much better in the long run.”

Pounds was very pleased with Makonnen stepping up to lead the BioHackathon project. “Every leader must eventually lead people with more expertise than themselves,” he said. “That requires listening, understanding, processing and building consensus around a shared goal. I was confident that Nobel [Makonnen] could do these things; he needed to do this to gain that confidence for himself. His teammates were very supportive, highlighting the outstanding collaborative culture we have at St. Jude. I’m very proud of Nobel and his team and very appreciative to the BioHackathon organizers and the Office of Data Science for making this opportunity possible.”

Launching a career

For Makonnen, the ABDS program is a foundation for a career in biostatistics or bioinformatics. He plans to pursue a PhD, likely in one of those disciplines, then continue in academia or industry — wherever he can remain hands-on in research and make the greatest contribution.

“When I applied, I shared that my goal in life is to earn a Nobel Prize, like my namesake,” Makonnen explained. “While I still have the dream, during the program, I’ve realized that doing work like this, pediatric cancer research, is very important — it can help kids, and that’s fulfilling in and of itself.”

That sense of purpose is central to what he would share with students considering the program. So is the willingness to take on an opportunity before feeling completely ready.

“If you have a drive to make an impact in the community and you want to feel like what you’re doing matters, the St. Jude ABDS program is great to be a part of,” Makonnen said. “I love coming to work every single day, and like my BioHackathon experience, our mentors keep pushing us to take the opportunities that will help build our careers.”

Future data scientists interested in applying to the ABDS program can do so here.

About the author

Senior Scientific Writer

Alex Generous, PhD, is a Senior Scientific Writer in the Strategic Communications, Education and Outreach Department at St. Jude.

More Articles From Alex Generous

Related Posts

A new dose of certainty for childhood cancer care in Mongolia

Taking steps toward nursing excellence worldwide

Pharmacists close gaps in global cancer care

Stay ahead of the curve