Researchers at UBDS3 describe lessons learned from organizing education and training in wartime
War destroys laboratories and instruments. It also severs the human networks through which science reproduces itself. It disrupts mentorship, collaborations and informal ties that turn students into researchers.
In a new Comment in Nature Genetics, an international team including Centre for Genomic Regulation (CRG) researcher Roderic Guigó argues that protecting science under the systemic stress of wartime means treating education and training itself as infrastructure, rather than as a series of one-off events.
The authors draw on the experience of running the Ukrainian Biological Data Science Summer School (UBDS3), held in person at Uzhhorod National University in western Ukraine since 2023, which is in session again this month from 11 to 25 July 2026.
Because martial law prevents most male students and early-career researchers from leaving the country, any model that depends on travelling abroad excludes them entirely. BDS³ inverts this by bringing world-class researchers into Ukraine, giving participants a point of contact with the international scientific community they would otherwise have no easy way to directly interact with.
What started as a two-week pilot that drew more than a hundred students to Uzhhorod, Ukraine’s westernmost city, UBDS3 has run every summer since, offered free of charge, with travel, room and board covered, and taught bilingually in Ukrainian and English by faculty from leading research institutions across Europe, North America and Asia, including the CRG.
The school is one half of a larger effort centred on Uzhhorod. A new genomics research facility in the city’s university has already begun to deliver clinical dividends, including corrected diagnoses for patients with a rare, treatable form of diabetes, a reminder that research depends entirely on the trained people the school exists to produce.
“Biological data science did not exist as an organised field in Ukraine before this initiative. My objective is to bring as much expertise to Uzhhorod as I can in whatever time I have left and turn it into the Cambridge of Ukraine,” says Taras Oleksyk, corresponding author from Uzhhorod National University and Oakland University in the United States and one of the main instigators of the school.
From this experience the authors set out three design principles for resilient scientific capacity. The first is a decentralised, cloud-based computing that does not depend on any single facility. Research institutes face infrastructure damage, energy instability and power cuts, so decentralised workflows can keep running even when local conditions are disrupted by events like air raids.
The second is a “train-the-trainer” structure in which former students return as instructors. Alumni now regularly return to teach the next cohort of BDS³, a self-renewing pipeline that has turned a series of courses into lasting institutional memory.
“It is an irreplaceable experience for the trainers. It is a sobering reminder of both how privileged we are and how fragile the stability of the world we live in is,” says Dr. Guigó.
Finally, the authors call for a non-extractive type of international collaboration, in which partners contribute expertise and access without recruiting participants permanently away so that internationalisation strengthens local science rather than draining it.
The school’s scientific leadership also includes CRG alumnus Fyodor Kondrashov, a former group leader now at the Okinawa Institute of Science and Technology, where directs the School of Molecular and Theoretical Biology. Working with Oleksyk and thanks to the support of the Zimin Foundation and the European Molecular Biology Organization (EMBO), he has helped establish the school. The CRG’s Roderic Guigó has also helped in the organization of BDS³ and has taught at past editions in person.
The lesson, the authors argue, is not confined to Ukraine. Climate disasters, energy failures, cyberattacks and funding shocks increasingly disrupt research systems everywhere, and the same distributed, people-centred model applies wherever infrastructure or mobility cannot be taken for granted.