Writing the detailed project proposal in 2023 proved invaluable for keeping me aligned with deliverables and objectives. At the same time, I quickly realized that adjustments were inevitable, whether to overcome data limitations, refine methods after conferences and trainings, or incorporate insights from the latest literature.

I spent the early months focusing on describing preterm birth (22-37 gestational weeks) trends and spatial patterns. We have an agreement with the region to access population-level health data covering deliveries from women resident in Lombardy between 2013 and 2024 (≈900,000 pregnancies). Unlike many studies that rely on single or few hospital data, we have population-level data, making our findings more robust and representative. The dataset also includes information to categorise preterm births by labour onset: spontaneous (labour starts naturally) and indicated (medically planned for maternal or fetal health).

To account for differences in population size across provinces (for example, Milan has far more births than smaller provinces) and make trends comparable over time, I calculated preterm birth rates by dividing the number of preterm births by total monthly deliveries for each labour category. At this stage, I could observe that spontaneous preterm births have been decreasing, while medically indicated preterm births followed a less clear, non-linear pattern. To gain a finer understanding, I grouped the data by the 12 provinces of Lombardy and used a two-week resolution, a compromise between weekly and monthly aggregation that retained enough observations to reliably compute trends.

Initial visualizations using rolling means revealed considerable variation, with preterm birth rates fluctuating by several percentage points (often around 2-6%) across provinces and labour types, which made me even more curious about potential environmental influences. Geographic and temporal differences often provide clues about environmental signals in the data. At this point, I realized that fully understanding these patterns would require exploring several modelling approaches, weighing the strengths and limitations of each before deciding on the best method to prepare for future Interrupted Time Series (ITS) analyses to investigate potential causal effects of environmental exposures on preterm birth. Below is a summary of the approaches I considered and what each was missing, which took a few months to learn:

For readers unfamiliar with Dynamic GAMs, Nicholas Clark’s mvgam package and blog are a well-documented starting point: https://nicholasjclark.github.io/mvgam/.

Finding Dynamic GAMs was a turning point. They allowed me to start describing both the temporal evolution and geographic variation of preterm birth rates in much more detail than before, capturing and quantifying subtle trends, differences between provinces, and patterns by labour type. That said, there is still work to do: I need more time to consolidate the results, fine-tune the model parameters, and carefully interpret the outputs before moving on to ITS analyses.

On 26 September 2025, this special night was celebrated all across Europe. More than 1.5 million people took part in activities spread across 460 cities in 25 countries. From experiments and hands-on games to lively discussions and exhibitions, it was a chance for people of all ages to dive into the world of science and connect directly with researchers in their communities.

This was my second time joining Researchers’ Night in Milan after my first experience in 2025, but this year felt different. For the first time, I was going to showcase the project I’m working on as part of my MSCA postdoctoral fellowship. My research is about figuring out what environmental factors might cause babies to be born too early, before 37 weeks instead of the full-term 40. What made it feel especially personal was that I’m using data from all pregnant women living in Lombardia. In a way, I was meeting people at the event who, until now, only existed in my dataset as numbers. Even though our dataset is fully anonymised, standing there and talking to people who might one day benefit from my work was both nerve-wracking and incredibly exciting. It made me feel the real impact my research could have.

In past editions, I had developed interactive games that people could play on their phones by scanning a QR code. This time, though, I wanted to create something hands-on that families could play together. It was time for some brainstorming with the young researchers from the Zuccolo group.

We sat down with Valeria and Alice and started listing concepts from my project alonside games we loved playing as kids (and maybe still enjoy now). Eventually, we settled on a Memory game, where players try to find matching pairs of cards, flipping only two cards at a time. The idea was to draw concepts from the project on the cards so we could explain them to participants as they played, and leave room to improvise or adapt the game depending on the players’ age.

However, our first attempt at a vertical poster for the game didn’t go as planned. We had used velkro to attach the cards but flippling one often pulled the whole poster down, definitely not kid-friendly. Then, Lucia suggested using wooden clothespins, and it worked like a charm. I also want to thank the Human Technopole communication team for suggesting a fabric poster, it not only looked good but could also be reused for future events. With that, we were ready for the night.

Running the activity throughout the evening was a learning experience in itself. alking to children and families was easy and fun, while discussing the project with curious university professors could be more challenging, especially when they came with preconceptions. We found the game worked best when we shortened it slightly, uncovering some cards to keep the pace lively. What impressed me most was how genuinely interested people were, teenagers included! Many not only played along but also contributed thoughtful ideas. We even gave participants the chance to draw their own cards, creating new concepts for us to explore.

I also want to give a big thank you to Laura and Giulia, who stepped in at the last minute to help me run the game. Having them there made the whole activity much smoother and, honestly, way more fun. As you can see, so many people contributed to this event, and it really reminded me how much collaboration and shared energy can bring a project to life.

For me, the whole evening was more than just outreach. It was also a reminder of why I love being a researcher. Walking around and seeing so many projects about water quality, forensics, wool, chessnuts, nuclear fusion, migratory birds, light, microscopes and many more; meeting other passionate researchers, and sharing ideas left me feeling recharged and inspired. It was like plugging back into the energy that keeps science alive.

I have also come away convinced that this memory-game format could be adapted to almost any scientific project. It’s simple, interactive and fun, perfect for bringing abstract concepts to life. So if you’re a fellow researcher looking for a way to engage the public, feel free to borrow the idea!

While I was drawn to the group’s vision for child and maternal health, I still had to consider whether HT was the right fit for me. I had already left my beloved Barcelona to pursue a PhD in Dublin, and the idea of taking another leap was not easy as this time. This job meant being in a country known for less favourable research contracts compared to the rest of Europe. I reached out to some Italian friends and colleagues to get a better sense of HT’s reputation. They had all heard about the institute (this was 2022). Yet, none of my Spanish or Irish friends had heard of it. HT had been making headlines in Italy for years as the country’s largest research investment. Friends (and friends-of-friends) reassured me that if I wanted to do research in Italy, HT was definitely my best bet.

Unlike other research centres and universities where most researchers were Italian, HT had a more international environment – 40% of its researchers came from abroad, half of them were Italians coming back after study or work in foreign institutions. On top of that, they offered four-year contracts (now extended to five years), which was unheard of for postdocs in Dublin or Barcelona. The salary was competitive, and under a new Italian law designed to attract research talent, 90% of my earnings would be tax-exempt.

So far, I had considered the research motivation and economic aspects, but what if the institute had a narrow, discipline-specific focus approach to science? I wanted to develop my interdisciplinary profile by tackling complex problems from multiple angles. Fortunately, HT takes a global and interdisciplinary approach to the study of human biology, integrating experimental and computational research. The institude brings together researchers from genomics, neurogenomics, structural biology, computational biology, health data science, molecular cell biology and biophysical modelling and simulation to study fundamental mechanisms of biological systems relevant to human diseases.

This was the collaborative environment I was looking for, where different fields intersect to push scientific boundaries. I am glad I decided to join Human Technopole in June 2023. Since then, I got the opportunity to attend both internal and external weekly seminar on topics beyond my expertise, broadening my perspective on research in unexpected ways. I also appreciate being part of a diverse postdoc community that promotes interdisciplinary approaches. Overall, I would encourage PhD students looking for postdoc positions to thoroughly research the group and institute to make sure their scientific goals and contract terms are a good fit.

We had no previous successful applications in the research center and I did not personally know anyone that had won an MSCA (Marie Skłodowska-Curie Actions) postdoctoral fellowship. I started by asking previous winners for guidance based on their published abstracts from the EU portal (Tip #1). I was surprised by how many replies I got, and I was soon meeting with a dozen of them, some of which kindly shared their winning proposals with me. This was a tipping point as it helped me understand the importance of structure, content balance, and importance of interdisciplinarity for this type of grant (Tip #2). They also emphasised that a key aspect of the proposal is to show that you are exactly the right person for the specific project based on your expertise, institution/research group that you will be working on, and the training that you could access to advance your future career (Tip #3). At that point, I was ready to start writing this proposal. However, the situation suddenly changed.

I was forced to change my initial idea for the project after a couple months working on it. I realised that the environmental data I wanted to use was lacking a key temporal characteristic. Learning from this, I believe conducting preliminary analyses early on is fundamental to understand the feasibility of an idea (Tip #4), as well as, seeking input from others early in the process (Tip #5).

In the following days, I presented a new project idea to both my supervisor (Dr. Luisa Zuccolo) and the Health data Science centre’s project manager (Dr. Giulia Mangiameli), and we agreed that it was worth the try despite the time constrains. The second round of proposal writing was easier despite still being stressful and challenging. We worked together with Luisa complementing each other’s skills and disciplinary background to craft this interdisciplinary project that combines data science, environmental science, perinatal epidemiology, and public health. After feedback a feedback session with my collegues the proposal was ready for submission (Tip #6).

Bonus tips from Dr. Carlos Jiménez

Carlos Jiménez is a postdoctoral fellow in Magda Bienko’s lab at Human Technopole, who developing proteomic-based tools to study spatial otganization in the cell nucleus. Check his MSCA project called PRUNE!

Writing a MSCA postdoctoral fellowship is a challenging and exhausting task, but also a priceless training for any future grant application. Indeed, I think that applying for this kind of fellowships is the best way to start a postdoc position after your PhD, even if you do not succeed, because it makes you get in touch with knowledge and concepts that may be necessary in your new position, catch up with the state-of-the-art in your field, and, most importantly, conceptualize and design the objectives and methodology of your research project for the following years.

My advice is to focus on the soundness of your research project before anything else: be sure to have clear and consistent objectives, to understand them, and provide a detailed methodology that proves how feasible they are. If you can, look for advice from MSCA alumni you have contact with, and carefully read and follow the suggestions from the handbook provided by the MSCA-NET. Finally, and speaking from my experience, if you fail in your first attempt, and you have the chance to reapply the following year, do not give up, use all what you learnt and the feedback from the evaluators, and try again: the chances are much higher in a second round.

Article about MSCA postdoctoral applications

Baumert P, Cenni F, Antonkine ML (2022) Ten simple rules for a successful EU Marie Skłodowska-Curie Actions Postdoctoral (MSCA) fellowship application. PLoS Comput Biol 18(8): e1010371. https://doi.org/10.1371/journal.pcbi.1010371

Chapter 1: Nanotechnology

Fresh out of school, I wanted to study life sciences in University but I was unsure of which one to choose. I was a good student in school but not an excellent one, which made difficult getting into medicine in Barcelona. When trying to select the next best option, I stumbled upon Nanotechnology (what a cool name I thought!). Nanotechnology involves studying and manipulating materials at the nanoscale, combining my favourite subjects of physics, chemistry, biology, and math to create advanced technologies in various fields. I ended up enjoying this degree, and I would do it again in a heartbeat, for both the degree and the people I met along the way.

The most valuable takeaway from this degree was learning to solve complex problems within interdisciplinary teams. I gained the ability to understand colleagues from diverse backgrounds, even without mastering a single discipline – just their interactions. However, I felt I lacked experience in optimizing parameters, making data-driven decisions, and working with large datasets to model outcomes.

Chapter 2: Data modelling

I decided to take a couple of classes on environmental modelling during my Erasmus in Ghent to overcome my lack of data modelling skills and … I loved it! We were able to simplify a system and even do some predictions at the click of a button. No need to process 100 samples to find a good enough reagent while wearing a lab coat or these funny plastic gloves. I was in for a treat. I rapidly start catching up with coding in R and enrolled in data modelling master’s also in Barcelona. I was the only one from Nanotechnology that had chosen this route as most of my friends decided to do material sciences, biomedicine, biotechnology or take a gap year.

During my master’s thesis, I had the chance to work with one of the top Climate and Health teams at ISGlobal in Barcelona. My supervisors, Xavier Rodó and Silvia Borràs, were incredibly passionate about research, and the enthusiasm was contagious. We investigated whether the composition of air masses could be linked to a rise in Kawasaki Disease cases in Japan. That’s when it clicked – I knew I wanted to do everything I could to pursue a career in research.

Chapter 3: Computer Science

When I first applied for a PhD position in Dublin through the Marie Skłodowska-Curie Innovative Training Networks (ITN), I wasn’t sure what to expect. The first interview didn’t go well as I struggled with the statistical questions, and nerves got the best of me. I left feeling disappointed, convinced I had missed my chance. But then something unexpected happened. One of the interviewers, Declan O’Sullivan, saw potential in me and suggested I apply for another position in Dublin. This position was in health data linkage using Semantic Web technologies, a field I knew nothing about.

People always say that choosing the right PhD supervisor is just as important as choosing the topic itself. I knew he was one of the good ones after having a Skype chat together. This was one of the best decisions I ever made!

The ITN provided me the opportunity to work with leading research teams in the health domain: epidemiology (Climate and Health group in ISGlobal Barcelona), rare disease research (Rare Kidney Disease group in Trinity College Dublin), health data governance (European Institute for Innovation through Health Data in Belgium), health data linkage (ADAPT centre in Trinity College Dublin). Even though COVID forced us to work remotely for most of my PhD, I still enjoyed being part of a network of PhD researchers, each tackling vasculitis from a different discipline or perspective.

Chapter 4: Health Data Science

After my PhD, I joined the Human Technopole in Milan (Italy) as a postdoc, working on child and maternal health with Luisa Zuccolo’s group. In the next posts, I’ll share more about this chapter, including how I went on to secure a MSCA Postdoctoral Fellowship within a few months of joining in.