One of the standout benefits of being part of MLGH—or any large research network—is the constant flow of opportunities shared by fellow academics. You’ll often see posts that say, “I’ve just found this event; please share it around, it looks promising!” When an opportunity is both relevant and timely, it’s the perfect moment to discuss it with your PI and make plans to participate.
That’s exactly what we did. I found myself on a flight to Hong Kong, with my research poster securely stored (in a fire equipment closet, ), heading to the Virus Evolution and Epidemiology summer school organized by the Croucher Foundation. The Croucher Foundation is an independent organization dedicated to advancing the natural sciences, technology, and medicine in Hong Kong by hosting workshops and events for researchers. A quick visit to their Projects and Events page showcases the wide range and high quality of their offerings. In my case, the summer school was hosted by the University of Hong Kong, although the Chinese University of Hong Kong also supports these initiatives.
The purpose of this course is best described as what is stated on their course page:
To provide practical training in the state-of-the-art computational analysis of the virus genomes for understanding their diversity, how they emerge, transmit, and evolve, and how they adapt to survive through acquiring resistance to drugs and immunity.
Above all else from this course, I was eager to learn how to properly make use of phylogenetic inference tools. While model setups from previous research papers—such as priors and clock models—can serve as valuable starting points, the real challenge lies in adapting them to suit our specific research question at hand. This requires not only a deep understanding of the underlying theory but also a collaborative approach, where ideas are exchanged and refined.
Making friends!
On the first day of our course, BEAST X was also launched, becoming the basis of our updated phylogenetic analyses. It’s important to distinguish BEAST X from BEAST 2, an independent project that employs random Markov Chain Monte Carlo (MCMC) instead of Hamiltonian Monte Carlo (HMC). The primary difference between these methods lies in how they propose samples: MCMC utilizes a random walk approach, whereas HMC leverages the gradients of the log-probability (a helpful primer here). HMC generally outperforms traditional MCMC in high-dimensional problems by avoiding the random walk behavior typical of simpler MCMC techniques. This advantage is crucial when working with large genetic datasets to infer reproduction numbers, mutation rates, and times to the most recent common ancestors (MRCAs), as these tasks involve navigating a larger dimensional space.
To make the most of a tool like BEAST X, it is essential to grasp its motivations and components. Our first day was dedicated to understanding phylogenetics—exploring why we study it and the methodologies we employ. The course brought together a diverse group of participants from Hong Kong, China, the USA, Italy, and Ireland (yours truly), encompassing various research backgrounds and areas of expertise, including epidemiology, biology, and bioinformatics (truly yours). This diversity fostered a collaborative learning environment where everyone could start from the basics and progressively build their knowledge together, ensuring that all participants, regardless of their prior experience, leave with new and applicable methods for their analyses.
The course content over the week was plentiful, and while I went in eager to absorb everything, ultimately I had my “highlighted” topics. Among those was phylogeography, particularly because it tied closely to the project I’ve been working on, which aims to identify transmission patterns internationally and within Singapore. But I can recall a number of “lightbulb” moments from subjects I lacked experience in, thinking to myself how this concept or that approach could be incorporated in my research. And with every module, the subsequent hands-on sessions really brought the theory into focus, making the theoretical feel tangible. The workshop tutorials can be found on the left panel of the BEAST website.
I'm known for my hand gesticulations.
The instructors were some of the most cited and respected figures in the field of phylogeny. So to those that are engaged with the school, this was an environment conducive to edifying discussions about halts in progress and potential solutions. Whether it was a technical issue with analysing large datasets or conceptual challenges in interpreting results, there was always someone with a fresh perspective or a novel solution. I found, in our one-to-one conversations, their concluding statement "If you have any questions, email me" to be especially encouraging, as if the importance and interest of our project was identified by others.
My first experience in participating in a poster was a humbling one. I’ve always been put off by posters overloaded with text and tiny, nearly indistinguishable graphs. So, in preparing my own, I took the opposite approach, opting for simplicity over volume. My gamble paid off to some extent—people stopped to look, asked thoughtful questions—but it also became clear that simplicity comes with its own set of challenges – namely, the difficulty to interpret with a dearth of information. I found that the true measure of a good poster is how well it conveys its message without requiring you to stand by and explain it. Balancing depth with accessibility is something I am still learning!
For those who had traveled far or were experiencing Hong Kong for the first time, the organisers thoughtfully wove in opportunities to engage with the local culture. My personal highlight came on the penultimate night when we boarded a cruise along the iconic Victoria Harbour. As stunning as the skyline was with its dazzling lights reflecting off the water, I remained down in the lower deck watching the academic façade from the week's classes be pushed aside in the face of a dance floor and an open bar! Nothing like the Macarena to lose all inhibitions.
I left Hong Kong with – apart from gifts for the family and hotel stationery – a greater sense of direction for the project and its potential. Thank you to all the organisers, the Croucher Foundation, and Swapnil Mishra for facilitating this!
A view from the upper deck. Alone up there before the lights display, and away from the karaoke below deck!
Enjoy Reading This Article?
Here are some more articles you might like to read next:
