Mapping groundwater aquifers located more than 400 metres beneath the seabed in the coastal area between Jesolo and Lignano. This was the objective of the recently completed oceanographic expedition carried out under the RESCUE project aboard the Laura Bassi research vessel of the National Institute of Oceanography and Applied Geophysics (OGS).

Coordinated by the University of Trieste, with OGS as a project partner, RESCUE aims to strengthen water security through research and innovation by identifying freshwater resources in unexplored coastal and offshore aquifers. Building on recent scientific discoveries, the project is mapping deep, low-salinity aquifers—located at depths exceeding 400 metres—both on land and beneath the seabed. At a time when freshwater resources in coastal regions are increasingly threatened by population growth, pollution and climate change, these aquifers could help supplement water supplies for drinking, agriculture and industry, while also reducing the costs associated with seawater desalination.

Researchers and students from OGS, the University of Trieste, the University of Pisa, CNR-ISMAR (Institute of Marine Sciences of the Italian National Research Council) in Bologna, the University of Malta, and the Monterey Bay Aquarium Research Institute (MBARI, California, USA) took part in the expedition aboard the Laura Bassi.

During the campaign, researchers carried out seismic reflection surveys to obtain high-resolution images of the subsurface beneath the seabed, together with seafloor mapping. In addition, electromagnetic surveys were conducted for the first time in the area to detect the possible presence of freshwater aquifers beneath the seabed. The investigations, which did not involve any seabed sampling, covered an area approximately 10 kilometres offshore, extending from Jesolo, in the Province of Venice, to the Tagliamento Delta off the coast of Lignano.

"The surveys were carried out under excellent environmental conditions, and all operations proceeded according to plan. The data collected are of very high quality, but several months of laboratory processing and analysis will be required before we can confirm the presence of freshwater aquifers beneath the northern Adriatic seabed," said Cristina Corradin, OGS researcher and scientific coordinator of the expedition.

"The operations required the integration of several geophysical techniques and close collaboration among the different research teams, the technical and scientific staff, and the ship's crew," explained Daniela Accettella of OGS, chief scientist aboard the Laura Bassi.

"After two years of scientific work, the RESCUE team has successfully applied innovative integrated geophysical methods for investigating deep offshore aquifers in one of the most promising study areas: the seabed of the northern Adriatic, particularly off the coast of Friuli Venezia Giulia. The new data will improve our understanding of regional groundwater resources and contribute to addressing water shortages linked to climate change and extreme events, such as the heatwave and drought currently affecting Europe," said Professor Michele Pipan of the University of Trieste, coordinator of the RESCUE project.

This oceanographic expedition has also laid the groundwork for a second research initiative funded by the European intergovernmental association JPI Oceans. Launched through a joint initiative by Malta and Italy, the new project will seek to confirm the presence of freshwater beneath the northern Adriatic seabed by drilling an offshore scientific borehole—the first of its kind in Europe. The project is coordinated by the University of Malta and involves OGS, the University of Trieste, the Alto Adriatico Technology Hub, as well as research institutions from Germany and Greece. It is funded by the Italian Ministry of University and Research (MUR) and the Autonomous Region of Friuli Venezia Giulia.

RESCUE – Resources in Coastal Groundwater Under Hydroclimatic Extremes is a project co-funded by the Water4All Partnership and the European Union. It addresses water scarcity in coastal regions by investigating deep groundwater resources located both onshore and beneath the seabed. The project focuses on mapping these resources, assessing their long-term sustainability, and developing policy recommendations for their sustainable use.

The project is coordinated by the University of Trieste, with OGS as a partner. Other partners include the University of Derby (United Kingdom), Ruden AS (Norway), and the University of Malta.

Professor Paolo Fornasiero, Full Professor of General and Inorganic Chemistry at the Department of Chemical and Pharmaceutical Sciences of the University of Trieste and Vice Rector for Research, has been appointed Fellow of the Italian Academy of Engineering and Technology (ITATEC) in recognition of his outstanding scientific achievements and his contributions to research in materials chemistry, catalysis and sustainable technologies.

Established in 2022 on the initiative and with the support of the Accademia Nazionale dei Lincei, ITATEC brings together leading experts in engineering, technology and applied sciences to promote scientific and technological excellence and strengthen the links between research, innovation and society.

The Academy fosters the exchange of knowledge among universities, research organisations and industry, develops evidence-based recommendations on strategic technological issues, provides independent scientific advice to public institutions and private stakeholders, and promotes the international visibility of Italian engineering and technological research.

At the European level, ITATEC serves as the Italian reference academy within the European Council of Academies of Applied Sciences, Technologies and Engineering (Euro-CASE), the network of national academies from 23 European countries that provides independent scientific and technological advice to European institutions and other international bodies.

Professor Fornasiero's appointment represents an important recognition of both his scientific leadership and the excellence of research carried out at the University of Trieste, reaffirming the University's commitment to advancing scientific knowledge, technological innovation and knowledge transfer at the national and international levels.

The University of Trieste Graduation Day returned to the heart of the city to celebrate research, knowledge and new generations.

On Thursday 16 July, Piazza Verdi hosted the proclamation ceremony for the PhD graduates of the 38th cycle, which concluded with the distinctive and traditional mortarboard toss. Included in the Trieste Estate programme, the event transformed one of the most significant rituals of academic life into a public occasion for bringing the University and the city community together.

Following its debut last year, Graduation Day in the city has confirmed itself as an event through which UniTS shares with the local community the value of doctoral education and the contribution of research to the cultural, scientific, social and economic development of society.

Rector Donata Vianelli’s first Graduation Day

For Rector Donata Vianelli, this was the first Graduation Day since taking office.

“This year’s ceremony was the first I experienced as Rector of the University of Trieste and represented the consolidation of a new tradition for the University and the city. PhD graduates stand at the highest level of university education and mark the entry of new generations into the research community. Celebrating this milestone in Piazza Verdi means bringing knowledge beyond the places where it is produced and sharing it with society, because science is not only the advancement of knowledge: it is responsibility, confidence in the future and the ability to contribute to the common good.”

Graduation Day 2026 in numbers

The 2026 edition confirmed the growth of the University of Trieste’s PhD programmes. The 187 new PhD graduates proclaimed in Piazza Verdi represent a new record number for the University.

As Francesco Longo, UniTS professor and Rector’s Delegate for PhD Programmes, pointed out, “this figure is also linked to the impact of scholarships funded by the PNRR, Italy’s National Recovery and Resilience Plan: the 38th cycle is in fact the first to reach proclamation after fully including the scholarships launched in the 2022/2023 academic year, alongside the University of Trieste’s decision to continue investing its own resources in doctoral education.”

The 2026 class of new PhD graduates is made up of 89 women, accounting for 47.6%, and 98 men, accounting for 52.4%. The international component represents 14.4% of the total, with PhD graduates from 14 countries: Austria, Brazil, China, Costa Rica, Croatia, Ethiopia, the Philippines, France, India, Iran, Nicaragua, Pakistan, Rwanda and Tunisia. Overall, including Italy, 15 nationalities are represented, confirming a doctoral community open to Europe, Asia, Africa and the Americas.

The 13 programmes: cross-disciplinary research

The 187 new PhD graduates come from 13 PhD programmes, reflecting the cross-disciplinary nature of research developed at the University of Trieste: from life sciences to physics, from chemistry to engineering, from artificial intelligence to nanotechnology, from sustainability to cognitive sciences, as well as history, architecture and territorial studies.

The programmes involved are: Environment and Life; Applied Data Science and Artificial Intelligence; Molecular Biomedicine; Chemistry; Circular Economy; Physics; Civil-Environmental Engineering and Architecture; Industrial and Information Engineering; Nanotechnology; Neuroscience and Cognitive Science; Reproduction and Developmental Sciences; Earth Science and Fluid Mechanics; History of Societies, Institutions and Thought. From the Middle Ages to the Contemporary Age.

Guests, institutions and partners

The special guest of the evening was Alberto Di Minin, Professor of Innovation Management at the Sant’Anna School of Advanced Studies in Pisa, researcher and science communicator, who delivered the keynote lecture “The Future Needs Teachers and Mentors. Reflections on Research, Open Innovation and Human Responsibility”.

His lecture focused on the role of research in major technological and social transformations. Every age has imagined the future through the science and technology of its time, but revolutions are never produced by technologies alone. They require people capable of turning knowledge into possibility, building bridges between research, business and society, and giving innovation a human meaning.

Alongside science, music also played a central role in the evening, thanks to the performance by Pierpaolo Foti, violinist and composer.

The ceremony, hosted by journalist Marinella Chirico, was attended by Rector of the University of Trieste Donata Vianelli, Rector’s Delegate for PhD Programmes Francesco Longo, President of the Autonomous Region of Friuli Venezia Giulia Massimiliano Fedriga and Mayor of Trieste Roberto Dipiazza. Alberta Gervasio, Chief Executive Officer of Bluenergy Group S.P.A., and Luca Cristoforetti, General Manager of CiviBank, also took part on behalf of the partners supporting the initiative.

The mortarboard toss

The evening culminated in the proclamation of the new PhD graduates through the ritual of the mortarboard toss. The gesture, one of the most recognisable images of graduation ceremonies, is commonly traced back to the U.S. Naval Academy: in 1912, new officers threw their old cadet caps into the air. Since then, the mortarboard toss has become an immediate symbol of transition and collective celebration.

The event was also made possible thanks to the support of CiviBank and Bluenergy Group S.P.A., local partners of the initiative.

Trieste Estate is organised and promoted by the Municipality of Trieste, with the support of PromoTurismoFVG and in collaboration with the Trieste Convention & Visitors Bureau.

A research team involving the European Laboratory for Non-Linear Spectroscopy (LENS), the CNR National Institute of Optics in Sesto Fiorentino, and the Universities of Florence, Trieste and Trento has developed an innovative technique based on the propagation of sound waves to study matter under extreme conditions. The results have been published in Nature Physics.

The researchers developed an innovative technique to investigate the behaviour of quantum matter by exploiting a familiar phenomenon: the Doppler effect. When an ambulance passes us, its siren sounds higher-pitched as it approaches and lower-pitched as it moves away.

“In the same way, sound travelling through a rotating superfluid changes depending on whether it moves in the same direction as the superfluid or in the opposite direction. By detecting these minute variations, the team was able to reveal the internal structure of superfluid Fermi gases, systems made up of atoms cooled to temperatures close to absolute zero. Under these conditions, Fermi particles pair up, giving rise to superfluidity, a state of matter in which friction disappears,” explain Marcia Fròmeta Fernandez and Diego Hernandez-Rajkov, CNR-INO researchers and lead authors of the study.

“We created a ring-shaped Fermi superfluid in which sound waves travel simultaneously in opposite directions. When the superfluid is set in rotation, the two sound waves are affected differently by the flow and accumulate a tiny difference in frequency: the Doppler effect.”

By measuring this effect with extreme precision, “we demonstrated that the circulation of the superfluid around the ring cannot take arbitrary values, but only quantised, or discrete, values, determined by the ratio between Planck’s constant and the mass of the particles. In this case, these values are half those expected for individual particles,” explains Francesco Scazza, Professor of Condensed Matter Physics at the University of Trieste.

“Our observation shows that the superfluid consists of fermions that pair up, forming composite quantum units with twice the mass, known as Cooper pairs, similarly to what occurs in superconductors. The result confirms that, even in these ultracold gaseous systems, superfluidity originates from the formation of fermion pairs,” concludes Giacomo Roati, Research Director at CNR-INO and head of the experimental group.

The study confirms the potential of atom interferometry, a technique that exploits the wave-like properties of atoms to perform extremely precise measurements. In this study, the technique was implemented using sound waves in the superfluid, proving to be an effective tool for investigating strongly correlated quantum systems.

One of the first results obtained was the observation that superfluidity progressively decreases as temperature rises. The work opens up a new avenue for studying the collective behaviour of particles and gaining a deeper understanding of the mechanisms underlying superfluidity and superconductivity, as well as other still-mysterious dynamics of quantum matter that are of fundamental importance for modern quantum technologies,” conclude Massimo Inguscio of LENS and Sandro Stringari of the Pitaevskii BEC Center at the University of Trento.

Their contributions to the development of atomic physics and their scientific collaboration laid the foundations for the work carried out at the LENS laboratory. The study comes in the year marking the centenary of the statistics formulated by Enrico Fermi in 1926 on the Arcetri hills, at the University of Florence, to describe the behaviour of particles such as electrons, protons and neutrons.

Which mammals live within the municipal territory of Trieste? Where are they concentrated? How numerous are they? For the first time, these questions can now be answered thanks to research coordinated by Alessio Mortelliti, Professor of Ecology at the Department of Life Sciences of the University of Trieste.

The study represents the first systematic survey ever conducted in the Municipality of Trieste on the presence, distribution and abundance of medium- and large-sized mammals. It fills a significant gap in knowledge about the city’s natural heritage and provides essential data for biodiversity conservation and land-use planning.

The research was carried out between November 2025 and February 2026 using a fully non-invasive methodology based on camera trapping. A total of 156 camera traps were installed across 78 monitoring sites distributed throughout the Municipality’s natural and peri-urban habitats, amounting to more than 2,300 observation nights. The cameras, automatically triggered by the passage of animals, made it possible to document wildlife presence without interfering with animal behaviour.

Overall, 18 mammal species were recorded, confirming the high naturalistic value of the Trieste area. Among the most widespread species were roe deer, wild boar and golden jackal. Of particular interest were the detections of the wildcat and European polecat, species of conservation relevance whose presence in an area so close to the city represents a finding of considerable scientific value.

In addition to compiling a checklist of the species present, the researchers used advanced statistical models to produce the first maps showing the probability of occurrence and abundance of the different species across the various environments of the municipal area, from forests and karst landscapes to agricultural and peri-urban zones. This information will make it possible to better understand the relationship between wildlife and habitat and to identify the areas where certain species are most concentrated.

“The results show that Trieste hosts a particularly rich and diverse mammal community,” explains Alessio Mortelliti. “This is not a phenomenon linked to climate change, but rather to the distinctive position of the Trieste area, which is closely connected from an ecological point of view with the Karst and the natural areas of Slovenia. This environmental continuity, together with the expansion of woodland across the Karst in recent decades, has favoured the presence of numerous wild species even close to the city.”

The data collected provide a valuable tool for updating knowledge of the fauna within the municipal territory and may support future activities related to biodiversity conservation, wildlife management, and urban and environmental planning by the Municipality of Trieste.

Identifying rare genetic diseases in the first days of life, before symptoms appear, and enabling more timely access to targeted care and clinical pathways: this is the perspective at the heart of the study conducted by the Medical Genetics group at the University of Trieste, led by Paolo Gasparini, Professor of Medical Genetics at UniTS and Director of the Medical Genetics Unit at the IRCCS Burlo Garofolo.

The work, carried out in collaboration with Stefania Zampieri, senior biologist in Medical Genetics at the IRCCS Burlo Garofolo, assessed the clinical and economic impact of integrating genomics into newborn screening programmes. The study takes Friuli Venezia Giulia as a model and compares the traditional approach with a “genomic-first” scenario based on whole-exome sequencing (WES) as the first-line investigation.

Newborn screening is now one of the most effective tools for prevention: it enables the early identification of a number of treatable conditions and allows interventions to take place before they can lead to severe complications, disability or irreversible damage. The traditional model, however, is largely based on the analysis of biochemical markers and can therefore detect only those diseases for which known and measurable signals are already available.

“Newborn screening is highly effective today, but it can identify only conditions associated with specific biomarkers,” explains Paolo Gasparini. “With a genomic approach, we can detect genetic diseases at their root, even in the absence of clear early signals.”

The difference concerns not only the technology used, but the entire diagnostic pathway. In the traditional model, screening begins with a biochemical test, followed, in the event of a positive result, by further genetic investigations where appropriate. In the genomic model, by contrast, whole-exome sequencing becomes the first step of the investigation, followed where necessary by targeted biochemical tests or additional diagnostic confirmation. In this way, genomics broadens the scope of screening to include conditions that are not currently covered by traditional panels.

The study examined the 2023 newborn cohort in Friuli Venezia Giulia, comprising 7,543 children. In that year, screening identified one case of spinal muscular atrophy, one case of cystic fibrosis and eleven cases of metabolic diseases. Applying the comparison between the two models to this cohort, the analysis estimates a direct cost of €131 per newborn for traditional screening and €183 per newborn for genomic screening, an increase of €51 per newborn.

Against this initial increase, however, the genomic model shows a positive cost-effectiveness profile. According to the estimates, in Friuli Venezia Giulia the “genomic-first” approach could generate overall savings of around €2.2 million, taking into account the economic benefits associated with earlier diagnosis, fewer complications and the reduction of longer and more complex care pathways. Genomic screening could also make it possible to identify a further 7–8 cases of rare diseases each year that cannot be diagnosed through traditional biochemical tests.

“Investing in early diagnosis means making the system more sustainable in the medium to long term,” adds Gasparini. “It is an approach to healthcare that is oriented towards prevention.”

The value of genomics applied to newborn screening is therefore measured not only by the number of conditions that can be identified, but also by its potential to reduce the so-called “diagnostic odyssey” that many families face before receiving a diagnosis. Earlier diagnosis can support timely access to more effective therapies, improve clinical and functional outcomes, reduce the care burden and contain the social impact of disability.

The study also underlines the need to define carefully which conditions should be included in any newborn genomic screening programme. The main criteria include the availability of a therapeutic intervention, the clinical validity of the test, age of disease onset, severity of the clinical presentation, penetrance and the technical feasibility of genetic analysis. The perspective, therefore, is not the indiscriminate use of genomics, but an application guided by clinical, scientific and ethical criteria.

Another central issue is the relationship between genomic screening and metabolic screening. The two approaches should not be regarded as alternatives, but as complementary. Genomics makes it possible to identify genetic variants associated with disease; metabolic screening can help validate their functional effect, distinguishing between pathogenic variants, findings of uncertain significance and different clinical forms, including late-onset forms.

The decreasing cost of sequencing, the growth of gene and cell therapies and the identification of new therapeutic targets are making genomics an increasingly important tool for preventive and personalised medicine. Key issues remain open, from the management of variants of uncertain significance to the protection of newborns’ genomic data, as well as the need for adequate bioinformatics infrastructure and pilot studies shared across laboratories and health systems.

Silvia Marchesan, Full Professor of Organic Chemistry at the Department of Chemical and Pharmaceutical Sciences (DSCF) of the University of Trieste, has been elected a Member of the Chemistry Division of the European Academy of Sciences (EurASc).

This prestigious recognition acknowledges Professor Marchesan's pioneering contributions to the field of nanomaterials science, particularly her research on the role of chirality in peptide self-assembly, which has provided fundamental insights for the design and development of functional peptide-based nanomaterials. Her election also recognizes her commitment to promoting interdisciplinary research, science communication, and dialogue across disciplines, cultures, and genders.

"I am deeply honoured to join the Chemistry Division of EurASc," said Silvia Marchesan. "This recognition marks an important milestone for my research group and for our commitment to building collaborations that transcend the boundaries between disciplines and scientific communities. I am delighted to contribute to the Academy's mission of promoting scientific excellence and international cooperation."

The European Academy of Sciences (EurASc) is an independent international organization that brings together some of Europe's most distinguished scientists and engineers to promote excellence in fundamental and interdisciplinary research, strengthen international scientific cooperation, and advance science and technology for the benefit of society. EurASc works alongside and complements the activities of national academies and the European Academies' Science Advisory Council (EASAC).

Professor Marchesan joins EurASc's Chemistry Division alongside several distinguished colleagues from the University of Trieste already serving in the Academy: Professors Paolo Fornasiero and Federico Rosei, members of the Materials Division from the Department of Chemical and Pharmaceutical Sciences, and Maurizio Prato, Professor Emeritus at the University of Trieste.

VIPoP – The Visual Politics of Populism, an ambitious research project funded under PRIN 2022 with a grant of €217,940, has recently come to a close. Over the course of twenty-nine months, the project investigated, from a comparative perspective, the visual communication strategies adopted by political parties in Europe. The project was led as Principal Investigator by Professor Mattia Zulianello, Associate Professor of Political Science at the Department of Political and Social Sciences of the University of Trieste. He coordinated a consortium of three research units, which also included the University of Milan and the University of Milano-Bicocca.

VIPoP represents the first systematic comparative analysis of visual populism in Europe and stands out for its deeply innovative approach, opening a new frontier in the field. For the first time, the ideational approach to populism was systematically extended to the visual dimension, which has so far remained largely marginal in the international literature. The project looks at political parties as genuine political brands, and at their symbols — logos, colour palettes and recurring visual codes — as identity markers governed by marketing logics not unlike those shaping competition among companies in the marketplace. From this perspective, VIPoP adopted a multi-platform research design, integrating thousands of data points from Facebook and Instagram and an advanced methodology combining computer vision, automated text analysis, qualitative interviews with party elites and randomised conjoint experiments.

“What struck us while working in the field was the level of awareness with which party communication teams manage every visual detail, much like a marketing department would handle its own brand identity. The interviews clearly showed that the choice of a colour, the framing of an image or the position of a logo are never random, but the result of a precise strategy. Reconstructing this visual grammar from a comparative European perspective was one of the most revealing aspects of the project,” Zulianello explains.

The relevance of VIPoP lies on a distinctly European scale. At a historical moment in which the populist challenge raises crucial questions about the resilience of the continent’s liberal democracies, the project provides essential analytical and interpretative tools for understanding how these political actors build consensus through images.

“Today, images are the primary vector of digital political communication: understanding their populist mechanisms is not an academic exercise, but a prerequisite for democratic citizenship. VIPoP therefore responds to Goal 16 of the 2030 Agenda — peace, justice and strong institutions — by providing tools to defend the quality of public debate and the resilience of European democracies,” Zulianello underlines.

Professor Zulianello’s scientific output within VIPoP has been particularly extensive and has appeared in top-tier international journals. Among the project’s most significant results is PopulisTree, a systematic mapping of European populist parties from 1979 to the present day, accompanied by openly accessible datasets covering national and European elections. Developed by Professor Zulianello and presented in an article published in European Union Politics, PopulisTree forms the classificatory backbone of the entire project and is intended as a reference tool for the international scientific community, as well as an open resource for anyone wishing to study the phenomenon with methodological rigour.

Other publications by Professor Zulianello connected to the project include a state-of-the-art review of populist visual communication in Political Studies Review, co-authored with Francesco Melito, a research fellow recruited within the framework of VIPoP; a study on the logos of populist radical right parties as elements of brand identity in The International Journal of Press/Politics, co-authored with Luigi Curini and Benjamin Moffitt; an analysis of perceptions of the mainstreaming of the populist radical right in South European Society and Politics, co-authored with Antonella Seddone; and the article “Show, Don’t Tell”, published in Political Studies in 2026 and also co-authored with Melito, which emerged directly from fieldwork interviews with the communication teams of Italy’s main political parties.

These contributions are accompanied by two books co-authored with Petra Guasti: Capire il Populismo, published by UTET in 2024, and Understanding Populism, forthcoming with Karolinum Press / University of Chicago Press.

The European Space Agency (ESA) has presented the largest and most detailed image ever produced of the centre of the Milky Way, the so-called Galactic Bulge: an extremely bright region densely populated with stars. This extraordinary “photograph” opens up new possibilities for scientists, who will be able to confirm the possible existence of exoplanets in this region and measure their mass through the tiny variations in starlight over time.

The Euclid Space Telescope acquired this enormous image in around 26 hours. It is a mosaic made up of nine pointings of its visible-light camera, each covering a portion of the sky larger than the full Moon.

In this image, Euclid captured more than 60 million stars, together with nebulae and star clusters. This extremely populated region of our galaxy is an ideal environment for the search for exoplanets through microlensing.

The news from Euclid is extraordinary: mapping the Galactic Bulge with such precision opens up unique prospects,” commented Gabriele Cescutti, UniTS Professor of Stellar Astrophysics. “Although this specific observing campaign was designed to exploit microlensing and search for exoplanets, such a density of stellar data is also extremely valuable for our research lines at UniTS. In our Department of Physics, we have been working for years on chemical evolution and ‘galactic archaeology’. We use spectroscopic and chemical data from stars to reconstruct, through theoretical models, the early history, formation timescales and origin of the elements in the bulge and nucleus of the Milky Way. High-resolution mosaics, such as the one produced by Euclid, are fundamental to understanding the precise distribution and nature of these stellar populations.

The University of Trieste is responsible for the operations of the two scientific instruments at the heart of the Euclid mission: VIS (Visible Instrument) and NISP (Near Infrared Spectrometer Photometer). In detail, UniTS researchers hold responsibility and coordination roles in several Key Projects dedicated to the scientific exploitation of Data Release 1 (DR1), expected around mid-2027, which concerns the study of cosmology through the statistical properties of the distribution and evolution of galaxies. UniTS also contributes to the production of cosmological numerical simulations based on High Performance Computing methodologies.

Image credits: European Space Agency - ESA.

Ten million cells analysed in less than two hours, with memory usage approximately three times lower than the best existing tools and speeds up to forty times faster on the largest datasets compared with the best existing tools. This is the remarkable result achieved by a group of researchers from the University of Trieste, Area Science Park, SISSA and Human Technopole, who developed DEVIL (Differential Expression with Variational Inference Learning), a new high-performance computational tool. The work has been published in Nature Communications.

Understanding which genes are active in cells is one of the keys to understanding diseases and developing new therapies. Today, the most advanced technologies make it possible to measure gene activity in millions of cells from dozens or hundreds of patients, generating an unprecedented amount of data for biomedical research. This revolution, however, brings with it two major challenges: on the one hand, the risk of errors in data interpretation; on the other, the difficulty of analysing such large volumes of information.

The first challenge is computational: analysing millions of cells requires enormous computing power. Traditional methods are too slow and consume too much memory to handle these volumes: a bottleneck that risks undermining the advantages offered by new data collection technologies. The second challenge is statistical. Cells collected from the same patient resemble one another more than they resemble cells from different patients, because they share the same individual biology, the same environment and the same personal characteristics. Ignoring this fact — as many currently used tools do — can lead to distorted statistical conclusions, with the risk of identifying as “significant” cellular changes that are not actually significant, or, conversely, of missing real ones.

To address these two issues, the researchers, thanks to DEVIL, succeeded in combining statistical rigour and computational speed in an innovative way. From a computational perspective, DEVIL, which was also developed with the support of Fondazione AIRC, was designed to make efficient use of the most advanced parallel computing architectures typical of artificial intelligence. Moreover, DEVIL is not only faster, but also uses less memory — a far from secondary detail. This means that analyses previously reserved for major computing centres can now become accessible to smaller research infrastructures and laboratories. From a statistical perspective, DEVIL addresses the problem through a Bayesian approach that correctly accounts for the structure of the data, treating cells from the same patient as correlated and therefore separating differences between patients from genuine differences in cellular activity.

“This work would not have been possible without ORFEO, the Area Science Park data centre, recently upgraded thanks to funding from Italy’s National Recovery and Resilience Plan,” says Stefano Cozzini, Director of Area Science Park’s Research and Technological Innovation Institute. “The availability of latest-generation GPUs, characterised by extremely high computing performance, together with careful optimisation of the algorithms for this architecture, developed by our team, now makes it possible to use DEVIL to address and solve problems on a significantly larger scale. We are very satisfied: it is not often that one can rely on a team with such high-level expertise, capable of making the most of the resources acquired.”

“Differential expression, that is, the statistical analysis that identifies which genes are significantly more or less active across two or more different biological conditions,” explains Giulio Caravagna of the University of Trieste, “is a mature technology. However, the transition to single-cell analysis has introduced statistical and computational issues that make the integrated analysis of large patient cohorts complex. Our work was developed precisely to overcome this bottleneck, combining methodological innovation and high-performance computing in order to scale up to the analysis of millions of cells from hundreds of patients.”

“In the development of DEVIL, the synergy between classical and Bayesian statistical tools represents a key strength within the reference oncological literature,” says Leonardo Egidi of the University of Trieste, “and makes DEVIL an efficient computational protocol with a strong methodological characterisation. Future developments could involve spatio-temporal models for multiple patients and introduce further computational approximations based on theoretical properties that are currently under study: a valuable combination of statistical, computational and biological expertise.”

DEVIL was tested on two concrete biological case studies. In the first, focused on the identification of immune system cells, the tool proved more precise and specific in recognising relevant biological functions. In the second, concerning the ageing of human muscle tissue, it identified age-related transcriptional changes in a more stable and biologically grounded way, reducing noise and highlighting key processes for subsequent analyses.

DEVIL has been released as free and open-source software, available to laboratories and hospitals around the world, paving the way for a new generation of large-scale genomic analyses for the study of tumours, degenerative diseases and the development of personalised medicine.