Pensoft to co-publish Problems of Dental Medicine with Faculty of Dental Medicine at Medical University Sofia

By joining Pensoft, the journal will benefit from improved publishing infrastructure and increased visibility and discoverability for its publications.

A dental publication titled "Problems of Dental Medicine" featuring a gloved hand holding a dental X-ray.

Pensoft Publishers has partnered with the Faculty of Dental Medicine at Medical University Sofia to co-publish the peer-reviewed journal Problems of Dental Medicine.

Published continuously since 1973, the journal offers a venue for high-quality research in dental medicine, public dental health, and healthcare. It publishes research and review articles and short communications, as well as methods, data, and forum papers in a wide range of disciplines in dentistry and dental sciences. Thanks to support from Medical University Sofia, authors of Problems of Dental Medicine can publish their articles for free.

Along with its most recent articles, the journal’s backlist content is also available on its website.

In joining Pensoft, Problems of Dental Medicine will get access to improved publishing infrastructure and benefit from increased visibility and discoverability for its published research. The journal will use Pensoft’s self-developed platform, ARPHA – an end-to-end publishing solution that makes it easy for both humans and machines to access, cite, and reuse research. ARPHA streamlines publishing workflows, providing a single online ecosystem for the entire editorial process, from manuscript submission to peer review, editing, publication, and archiving.

“Problems of Dental Medicine has a long history and established traditions, but joining Pensoft marks a major step forward in the journal’s development,” said Editor-in-Chief Prof. Boyko Bonev. “I am confident that this collaboration will be extremely beneficial, providing broader international visibility, greater accessibility, and new opportunities for the advancement of dental science.”

“We are excited to start this new chapter with Problems of Dental Medicine and believe this partnership will significantly contribute to the development and understanding of dental science,” commented Prof. Lyubomir Penev, CEO and founder of Pensoft and ARPHA.

Bugs in drawers: entomological outreach collections and their role in community engagement

Pensoft’s new Diamond Open Access journal Natural History Collections and Museomics (NHCM) has just concluded its first topical collection, “Entomological Outreach Collections and Community Engagement.”

The collection brings together contributions from faculty, curators, collection managers, and students from around the world, having sprung from the November 2024 “Collections Reaching Out” symposium at the Entomological Collections Network annual meeting in Phoenix, Arizona. The symposium focused on the conception, preparation, management, and educational use of outreach drawers: special collections of insect specimens curated for public learning and awareness.

A tray of insect specimens surrounded by children drawing them.
A high school outreach class using display drawers. Credit: Abby Weber.

Despite their crucial role in science education and public engagement, outreach drawers have remained understudied. The papers in this new collection aim to fill that gap, examining their purpose, design, usage, and how to assess their impact. The editors, Victor Gonzalez and Jennifer C. Girón Duque, brought together international perspectives to identify best practices and strategies that maximise outreach effectiveness.

Featured in the collection

Two children and a teacher looking at a tray of insect specimens.

Children observing insects during National Science Week 2024. Image credit: Queen Victoria Museum and Art Gallery, Tasmania.
Various trays of insect specimens.

Educational collection of the Invertebrate Division at the LOUNAZ Museum. a Young visitors at an insect educational session; b insect cabinets and display boxes showcasing curated specimens. Credit: Paucar-Cabrera et al.
Painted models of insects, with people painting them.
 Painted 3D models along with the acrylic paints used during the European Researchers’ Night workshop. Credit: Albert et al.
A tray of insect specimens.
A Museums in Miniature drawer ready for deployment. Credit: Wooden et al.
A tray of bee specimens.
A display drawer of British bees, utilising unit trays system. Credit: Ashleigh L. Whiffin.
Close up of a green bee.
Augochlora pura (female). Credit: Anthony Abbate.

By exploring topics from innovative educational programs and accessible specimen displays, to community science initiatives and frameworks for assessment, the collection demonstrates how outreach collections can turn “bugs in drawers” into powerful vehicles for learning and community engagement.

NHCM’s Diamond Open Access model ensures that these resources are free and available to all, reaffirming the journal’s commitment to inclusivity and expanding the societal impact of natural history collections.

You can read the topical collection here.

The bee’s knees: a new, non-lethal way to study pollinator networks

Guest blog post by Alexander Edwards, lead author of the paper ‘Case study of non-lethal sampling for plant-pollinator networks via barcoding and metabarcoding on bumble bees in Germany‘ published in Metabarcoding and Metagenomics.

“Who are you and what have you done?”

That’s not just a question for crime shows. It’s also exactly what we want to ask every pollinating insect we catch. Who are you – which species? And what have you done – which flowers have you visited, carrying pollen from one to the next? And these questions are important!

Do you like chocolate? Or maybe coffee? How about apples, strawberries, or cherries? All of them need insect pollinators. Unfortunately, many insects are in decline, threatening the stability of ecosystems. Studying them is more important than ever. Traditionally, researchers relied on field observations of foraging behaviour. But these are time-consuming and can never capture the full picture. That’s why many groups now use genetic approaches, studying the pollen carried by insects. This method is called pollen-metabarcoding, where a short genetic sequence from a specific region (in our case, ITS2) can identify the plant species that pollen belongs to.

There’s just one catch: collecting pollen often means killing the insect. Lethal sampling does have its merits – the preserved specimen can be used to answer other questions in the future, for example. But in our case, it felt counterintuitive. We want to study species that might already be endangered, and killing them could worsen their situation. So we had to find another way.

 A man crouching in a field sampling a bee specimen.
In the field, we found queen-marker cages easy to use and really helpful to get
the genetic material we need for our analyses. Credit: Willi Müller.

Enter: the queen-marker cage. This tool – borrowed from beekeepers – is a plastic tube with a mesh at one end and a plunger at the other. We used it to capture a bumble bee, immobilize it, and remove its pollen. The mesh is also large enough to clip one of the bee’s feet – in insects, this part is called the tarsus. That may sound harsh, but it doesn’t significantly affect the animal. During our collection, we observed those five-footed bumble bees visiting flowers and gathering pollen in the days after release, with no difference to their six-footed counterparts.

A bee in a cage trap with a mesh top, beneath a pair of scissors.
The mesh of the queen-marker cage is just big enough for a fine pair of scissors
to fit through the holes and allow for precise clipping of the bumble bee’s tarsus, here
Bombus lucorum agg.

But why take a bee’s foot in the first place? Many animal groups include cryptic species – species that look almost identical, even to experts. Bumble bees are no exception. The only reliable way to tell these look-alike species apart is through DNA barcoding, often using the COI gene. A single clipped foot provides just enough tissue to barcode the bee itself, in addition to analyzing the pollen it carried.

Of course, the work doesn’t end in the field. Once we had the samples, the real detective work started in the lab. And as everyone with hay fever knows: pollen gets everywhere. To avoid contamination, we processed everything in a special clean lab, wearing full-body protective suits. It may seem over the top, but when you’re working with invisible grains of pollen, even the smallest contamination can skew the results.

A man in protective clothing working in a lab.
As pollen is everywhere, we had to use a special clean lab with the
corresponding attire.

And in the end, it worked! Our results matched expectations: we detected a cryptic species (fittingly named Bombus cryptarum), we saw that the longest flowers were visited only by bumble bees with the longest tongues, and the pollen we identified came from plants flowering at our study sites during collection.

Now we have a simple and non-lethal way to gather the genetic material needed to identify pollinators and the flowers they visit, answering ‘Who are you, and what have you done?’ – without adding pressure to vulnerable insect populations.

Bees on purple flowers.
Studying pollination helps us understand and protect ecosystems. Bombus
lucorum
agg. and Apis mellifera on Phacelia tanacetifolia. Credit: Anna Wurster.

Orignal source

Edwards A, Gemeinholzer B (2025) Case study of non-lethal sampling for plant-pollinator networks via barcoding and metabarcoding on bumble bees in Germany. Metabarcoding and Metagenomics 9: e141904. https://doi.org/10.3897/mbmg.9.141904

Follow Metabarcoding and Metagenomics on Bluesky and Facebook.

Pensoft to co-host a session on knowledge transfer & continuity at Living Data 2025

Pensoft is a co-organiser of a four-hour session, titled: “Long Live Biodiversity Data: Knowledge Transfer and Continuity across Research Projects”. 

In October 2025, four major institutions in the biodiversity research landscape: TDWG, GBIF, OBIS and GEO BON, will come together as the organisers of the Living Data 2025 conference. 

The event is set to be among one of the most crucial international gatherings of the year for experts and stakeholders in the field of biodiversity data. Set to take place in the Colombian capital of Bogotá between 21st and 24th, Living Data 2025 will centre around four core themes:

  • Open data
  • Data integration
  • Biodiversity data application
  • Community engagement and capacity-building

As an academic publisher with experience and commitment to all these thematic areas, Pensoft will participate in the event in the capacity of an exhibitor and an award sponsor, as well as a symposium host. 

The conference delegates will have the chance to learn more about the publisher, its exclusively open-access scholarly portfolio and participation at various international scientific projects when they visit the company’s branded stand. 

During the event, the scientific publisher and technology provider will also present the Pensoft Award for the Best Student Oral Presentation, which grants the winner a free publication in an open-access, peer-reviewed journal from our portfolio

Crucially, Pensoft’s involvement in the Living Data 2025 programme also includes a dedicated four-hour session titledLong Live Biodiversity Data: Knowledge Transfer and Continuity across Research Projects”. 

The symposium will be jointly co-organised by Pensoft, LifeWatch ERIC and the Naturalis Biodiversity Centre. As the title suggests, the session will focus on the longevity of scientific outputs as they are generated, shared and re-used across disciplines, organisations and initiatives. In this context, tools, information hubs and workflows enabling exchanges that truly consolidate the global biodiversity data space over time will be showcased.

In a broader sense, the session will also seek to demonstrate how targeted communication can help transform science results into actionable knowledge by raising awareness among agenda-setters. This will speak to the potential of a multi-level approach to information sharing to bridge the gap between science and policy in relation to increasingly ambitious global environmental objectives.

Multiple projects affiliated with Pensoft will be represented in these deliberations, in order to share a diverse array of relevant insights:

The symposium will be divided into two sessions:

  • 22 October (Wednesday): 10:45 AM – 12:45 PM (UTC/GMT-5)
  • 23 October (Thursday): 10:45 AM to 12:45 PM (UTC/GMT-5)

You can find out more about Living Data, including the details on registering for an in-person or virtual attendance, on the conference’s website. Our session is listed on this page under ID number 6788879.

As an additional note, the organisers of the conference have launched a call for extended abstracts for all speakers at Living Data 2025 that will remain open until 1st October 2025. The participants who opt to publish their conference abstracts in the Biodiversity Information Science and Standards (BISS) journal will enjoy permanent and far-reaching accessibility and discoverability for their conference contributions.

The TDWG network, who launched BISS as their official scholarly outlet in 2017 in collaboration with long-time partner Pensoft, have posted a list of the advantages for submitting an extended abstract, even though they have already had their abstracts accepted by the Living Data 2025 organisers. Amongst the reaslons are many perks typically associated with a conventional research article, such as DOI registration, indexation at dozens of scientific databases, embedded media, tables and supplementary materials, and usage metrics.

Scientific Annals of the Danube Delta Institute publishes its first articles with Pensoft

Established in 1993, the journal offers a multidisciplinary platform for studies related to wetlands and deltaic environments.

The first articles of Scientific Annals of the Danube Delta Institute (SADDI) —the official journal of the Danube Delta National Institute for Research and Development in Tulcea, Romania—have now been published on the journal’s new website following its transition to Pensoft’s ARPHA platform.

Two white spoonbills standing in shallow water, reflecting in the pond, surrounded by lush greenery and scattered debris.
Eurasian Spoonbills (Platalea leucorodia) in the Gandoman Wetlands. Photo by Ruhollah Asgari

The newly published articles focus on supporting the sustainable management of the Danube Delta, including an assessment of the impact of pesticide residues on the river and some insights on the potential of earth observation data for monitoring the Delta. The journal also publishes research on the ecology and biogeography of wetland ecosystems in general, such as a study on the habitat and behaviour of the Eurasian Spoonbill (Platalea leucorodia).

Established in 1993, the journal offers a multidisciplinary platform for studies in the fields of ecology, biodiversity, environmental protection, hydrology, ichthyology, ornithology, limnology, and sustainable development related to wetlands and deltaic environments. It is free to publish and open for submissions.

“With the launch of the first articles in Volume 30 of the Scientific Annals of the Danube Delta Institute, now published with Pensoft, we celebrate an important milestone in advancing research on biodiversity, aquatic ecology, and data integration,” said the journal’s Editor-in-Chief Iuliana-MihaelaTudor of the Danube Delta National Institute for Research and Development.

“This new stage reflects not only the scientific value of the contributions, but also the collaborative spirit that strengthens our community and drives innovation forward.

“As Louis Pasteur once said, ‘Science knows no country, because knowledge belongs to humanity, and is the torch which illuminates the world.’

“We warmly invite our readers, authors, and collaborators to explore this volume, to share in the exchange of ideas, and to continue building together a future where open science connects and empowers us all.”

New study highlights critical communication challenges in European biodiversity research networks

The research uncovered issues that hinder scientific collaboration and data integration across Europe.

A new study published in June reveals that fragmented communication and coordination undermine the effectiveness of Europe’s leading biodiversity research infrastructures (BioRIs). The research led by Allan T. Souza, Tomáš Martinovič, Carrie Andrew, Yi-Ming Gan, and Erik Kusch, examined the communication strategies of four major European BioRIs, uncovering issues that hinder scientific collaboration and data integration across the continent.

BioRIs address the complex challenges in biodiversity research, scientific collaboration across disciplines and national boundaries, and inform the public and policymakers about the status and challenges of the European biodiversity. The new study focusing on the communication and coordination amongst BioRIs revealed issues in communication strategies of the key European BioRIs, including DiSSCo, eLTER, GBIF and LifeWatch ERIC. 

A graphic showing logos of biodiversity initiatives: GBIF, eLTER, DiSSCo, and LifeWatch ERIC arranged like puzzle pieces.
Conceptual representation of the Biodiversity Digital Twin (BioDT) with the integration of the four research infrastructures (DiSSCo, eLTER, GBIF and LifeWatch ERIC) involved in the development of the digital twin.

The study revealed uneven geographical representation, inconsistent communication practices, and limited data and service cohesion, ultimately impeding collaboration and efficient resource use. The distribution of BioRIs across Europe is imbalanced, leading to unequal research opportunities and capacities between regions. While communication within individual BioRIs is often strong, collaboration between infrastructures remains limited. The absence of shared tools and standard communication channels severely hampers joint efforts and information flow. The study found no standardised approach to communication methods among BioRIs, creating confusion for stakeholders and reducing transparency and accessibility for scientists, policymakers, and the public. Variations in data standards across infrastructures hinder data integration. While some infrastructures share common standards, others maintain highly specialised protocols, restricting broader collaboration and data use. Although some initiatives show promise for harmonisation, broader systemic challenges persist.

Maps of Europe showcasing various biodiversity projects: BIODT, DISSCO, eLTER RI, GBIF, and LifeWatch ERIC, highlighted in different colors.
Distribution and representation of European Biodiversity Research Infrastructures is fragmented (A), with each individual research infrastructure differing in its geographical coverage (B).

“Our findings highlight a critical need for a unified communication framework to break down barriers between these infrastructures,” said lead author Allan T. Souza. “Without it, we risk missing opportunities for impactful, cross-disciplinary research vital to tackling Europe’s and the world’s biodiversity challenges.” While some initiatives to tackle this issue demonstrate the potential for harmonisation, the broader systemic challenges persist.

Violin plot comparing collaboration intensity in research infrastructures (DiSSCo, eLTER, GBIF, LifeWatch) by network type (Between/Within).
Self-reported collaboration intensity within and between four biodiversity research infrastructures (BioRIs). Upper panel: Collaboration of all BioRIs (DiSSCo, eLTER, GBIF and LifeWatch) are largely siloed, with minimal interaction across BioRIs being reported. Lower panel: Across BioRI, collaboration is uniformly low (left), while within BioRI, collaboration varies, with LifeWatch and eLTER showing the highest intensity (right). These patterns highlight limited cross BioRI integration and variable internal cohesion, highlighting the fragmented scenario in the BioRI landscape.

To overcome these barriers, the study argues for these urgent steps. The first one recommends a standardised communication framework. Creating a shared platform with tools for chat, mailing, discussion boards, calendars, and clear public and internal information separation. Another recommendation is to increase geographical coverage that proposes long-term balancing of BioRI representation across Europe through coordinated funding and capacity sharing. The authors also suggest promoting data interoperability, while harmonising standards and deepening understanding of cross-domain differences to improve integration. And last but not least, it is important to leverage good practice examples.

Envisioned solution/tool architecture for the Biodiversity Research Infrastructures (BioRIs) communication and coordination, showing pages in columns and individual components of the solution/tool. Green components are publicly available, orange one are available to everyone after they login to the system and red ones are available only to the members of the given group. At the bottom are Github repositories where the relevant information for given web page are stored.

Addressing fragmentation within European BioRIs requires improving communication, coordination, and interoperability through both technical and institutional measures. Strategic funding, shared platforms, and community engagement will be key to building a more integrated and efficient research network. The study highlights that changes should be gradual, systematic, and informed by proven models of collaboration.

Research article:

Souza A, Martinovič T, Andrew C, Gan Y-M, Kusch E (2025) Fragmented Networks: Challenges in communication and cohesion of European Biodiversity Research Infrastructures. Biodiversity Data Journal 13: e148079. https://doi.org/10.3897/BDJ.13.e148079

After 150 years, butterfly collected by Alfred Russel Wallace finds its name

A monograph by a team of international researchers revised the diverse Neotropical butterfly genus Euptychia.

Guest blog post by Shinichi Nakahara

A new Amazonian butterfly

In a drawer of the butterfly collection at the Natural History Museum in London (NHMUK) sit a handful of rather unassuming butterfly specimens from Amazonia. Marked only with a couple of bands and eyespots on the underside of their wings, these specimens were thought to be pitch-brown black-eyed satyrs (Euptychia picea) for over 150 years. Labels attached to some of them suggest that they were collected by 19th-century naturalist Alfred Russel Wallace during his expedition to the Amazon rainforest along with his good friend, Henry Walter Bates. A recent 300-page monographic study of this group of butterflies has recognised these specimens as a species new to science: Euptychia andrewberryi, or Berry’s black-eyed satyr.

A monograph by a team of international researchers revised the diverse Neotropical butterfly genus Euptychia. The study was led by Shinichi Nakahara, a Lepidoptera subject editor for ZooKeys, Biodiversity Data Journal, and Neotropical Biology and Conservation.

The story behind the eponym

In 1848, Wallace and Bates set sail on an epic voyage for the Amazon to explore and research its neotropical biodiversity. Their trip was funded by Samuel Stevens, a natural history agent in London. In 1852, after four years of collecting, Wallace returned from Brazil with all his specimens loaded on a ship named the Helen. However, on his way back to London, the Helen caught fire in the middle of the Atlantic Ocean and sank. In the blink of an eye, Wallace lost all of his material. This shipwreck is one of the most disastrous events in the history of science, yet fortunately some of Wallace’s Amazonian material escaped the tragedy and made its way to London.

A black-and-white photo of an old gentleman in formal attire, sitting in an ornate chair at a photo studio.
Alfred Russel Wallace.

Throughout the course of his travels in Amazonia, Wallace occasionally shipped back specimens to Stevens in London. In a letter to Stevens, dated 15 November 1849, Wallace wrote from Santarém (500 miles above Pará), “I spent about three weeks at Montealegre and have now been back here nearly a month, so before I leave for the Rio Negro, I send you a small lot of insects; they consist almost entirely of Lepidoptera…

The registration records at the NHMUK suggest that the holotype of Euptychia andrewberryi entered the collection in 1850, supporting the notion that it was shipped to London while Wallace was still in Brazil.

A photo of a middle-aged man with short, light gray hair wearing black, seated in a vibrant café setting.
Andrew Berry. Photo by Shori Hijikata

Andrew J. Berry is a historian of science with a long-standing interest in Wallace. A lecturer in the Department of Organismic and Evolutionary Biology at Harvard University, Andrew teaches evolutionary biology and related subjects. While working on the Euptychia monograph, Andrew’s dry sense of humor and undying zest were critical in the completion of a study of such great magnitude. 

Why monographs are important

Taxonomic monographs serve as a foundation for understanding species diversity and phylogenetic relationships within a given group of organisms. They enhance our understanding of the myriad species with which we share our planet. Undertaking work on a monograph typically requires travelling to many museums, incorporating morphological and molecular studies, and conducting field work, as well as reviewing literature to solve taxonomic or nomenclatural conundrums. Despite this tedious process, which requires a great deal of perseverance, monographs underscore the rather underappreciated nature of taxonomy, representing “a small gain for a huge effort.” Unfortunately, most diverse tropical groups of organisms have not yet received a comprehensive taxonomic treatment. 

Euptychia is the largest Neotropical satyrine genus ever to be revised. This monograph increased the known diversity of Euptychia to 50 species, significantly more than hitherto estimated. Although further testing is needed, it is worth emphasising that only 16 described species were recognised in the genus in 2004. It is remarkable to realise that such a profound discrepancy existed between perceived and actual species richness, especially within a well-studied group of invertebrates such as butterflies. This further underscores the importance of revisionary monographic works.

Til moth do us part: new species marks 42 years of marriage

“It is without a doubt the prettiest species I have encountered in my long scientific career,” said Huemer, who named the moth after his wife.

European Lepidoptera (butterflies and moths) are generally considered well-known and thoroughly researched. Nevertheless, researchers discover new species every year; most of which are inconspicuous, so-called ‘cryptic’ species, previously overlooked.

Colourful species, on the other hand, have been largely catalogued in Europe as they attract a lot of attention, which made the surprise and delight at the discovery of an extraordinarily striking, and previously unnamed, moth all the greater.

Moth specimen.
Ingrid-Maria’s carcina (Carcina ingridmariae). Credit: Peter Huemer/Ferdinandeum.

A newly discovered, pink species has now been named Carcina ingridmariae by Peter Huemer, a scientist at the Tyrolean State Museum Ferdinanduem (Innsbruck, Austria). Huemer published his discovery in the open-access journal Alpine Entomology.

According to current knowledge, the moth is widespread in the eastern Mediterranean region: distributed from Croatia across large parts of Greece and Cyprus to Turkey. However, more detailed studies on its distribution are still pending.

Seemingly unmistakable

Ingrid-Maria’s carcina belongs to a species-poor group of butterflies. In Europe, there is only one other species of the same genus, the oak carcina (Carcina quercana).

Moth on a leaf.
The oak carcina (Carcina quercana). Credit: mazzeip via iNaturalist.

This widespread moth was described as early as 1775 by the famous naturalist Johann Christian Fabricius based on specimens from Saxony, and is distributed from North Africa across large parts of Europe to the Balkans.

Due to its unusual colour, the species has always been considered unmistakable. In fact, it is so popular even among amateur researchers that it adorns the cover of an important British identification book.

But, hiding in plain sight, was a second species, mistaken for the oak carcina for more than 100 years.

As a result of its apparent unmistakability, Carcina ingridmariae was always misidentified and was first published – incorrectly – as the oak wood carcina from Crete in 1916.

Mountain landscape.
Habitat of Carcina ingridmariae in north Cyprus (eastern part of Five Finger Mountains near Kantara). Credit: Peter Huemer/Ferdinandeum.

It was only the introduction of new molecular identification methods that put researchers at the Ferdinandeum on the trail of the nameless moth. DNA barcodes, also known as genetic fingerprints, showed huge differences of more than 6% between the two species.

Subsequent morphological examination of the sexual organs led to the famous “wow” effect. And, upon closer inspection, the two species could not be confused at all, despite the confusingly similar external appearance of the species: namely, a wingspan of about 2 centimeters, a pink base colour with yellow spots, and strikingly long antennae.

A special gift for a 42nd wedding anniversary

Peter Huemer has described more than 200 species from Europe in 35 years, but is particularly enthusiastic about this new species. He said: “It is without doubt the prettiest species I have encountered in my long scientific career, even though it was still unnamed.”

It was therefore obvious to Huemer that he should dedicate the new species to his wife, Ingrid Maria, on their 42nd wedding anniversary. The researcher justifies this choice of name above all with his wife’s decades of support for his work.

Original source

Huemer P (2025) The supposedly unmistakable mistaken: Carcina ingridmariae sp. nov., a surprising example of overlooked diversity from Europe and the Near East (Lepidoptera, Peleopodidae). Alpine Entomology 9: 51-63. https://doi.org/10.3897/alpento.9.158239

Follow Alpine Entomology on Bluesky and Facebook.

Ensuring drug safety using AI models for adverse drug reaction prediction

An AI model developed to predict adverse drug reactions could potentially support early-stage drug safety assessment before clinical trials.

Adverse drug reactions (ADRs) are a significant cause of hospital admissions and treatment discontinuation worldwide. Conventional approaches often fail to detect rare or delayed effects of medicinal products. In order to improve early detection, a research team from the Medical University of Sofia developed a deep learning model to predict the likelihood of ADRs based solely on a drug’s chemical structure.

The model was built using a neural network trained using reference pharmacovigilance data. Input features were derived from SMILES codes – a standard format representing molecular structure. Predictions were generated for six major ADRs: hepatotoxicity, nephrotoxicity, cardiotoxicity, neurotoxicity, hypertension, and photosensitivity.

A flowchart illustrating chemical transformation, featuring molecular structures, fragment analysis, and decomposition stages.
Visual representation of SMILES and the process of molecular deconstruction. Adapted from Wu JN, Wang T, Chen Y, Tang LJ, Wu HL, Yu RQ. t-SMILES: a fragment-based molecular representation framework for de novo ligand design. Nat Commun. 2024 Jun 11;15(1): 4993. https://doi.org/10.1038/s41467-024-49388-6.

“We could conclude that it successfully identified many expected reactions while producing relatively few false positives,” the researchers write in their paper published in the journal Pharmacia, concluding it “demonstrates acceptable accuracy in predicting ADRs.”

Infographic detailing an AI model predicting adverse drug reactions for various compounds.

Testing of the model with well-characterized drugs resulted in predictions consistent with known side-effect profiles. For example, it estimated a 94.06% probability of hepatotoxicity for erythromycin, 88.44% for nephrotoxicity and 75.8% for hypertension in cisplatin. Additionally, 22% photosensitivity was predicted for cisplatin, while 64.8% photosensitivity was estimated for the experimental compound ezeprogind. For enadoline, a novel molecule, the model returned low probability scores across all ADRs, suggesting minimal risk.

Notably, these results demonstrate the model’s potential as a decision-support tool in early-phase drug discovery and regulatory safety monitoring. The authors acknowledge that performance of the infrastructure could be further enhanced by incorporating factors such as dose levels and patient-specific parameters.

Research article:

Ruseva V, Dobrev S, Getova-Kolarova V, Peneva A, Getov I, Dimitrova M, Petkova V (2025) In situ development of an artificial intelligence (AI) model for early detection of adverse drug reactions (ADRs) to ensure drug safety. Pharmacia 72: 1–8. https://doi.org/10.3897/pharmacia.72.e160997

Mining nature’s knowledge: turning text into data

By using natural language processing, researchers created a reliable system that can automatically read and pull useful data from thousands of articles.

Guest blog post by Joseph Cornelius, Harald Detering, Oscar Lithgow-Serrano, Donat Agosti, Fabio Rinaldi, and Robert M Waterhouse

In a groundbreaking new study, scientists are using powerful computer tools to gather key information about arthropods—creatures like insects, spiders, and crustaceans—from the large and growing collection of scientific papers. The research focuses on finding details in published texts about how these animals live and interact with their environment. By using natural language processing (a type of artificial intelligence that helps computers understand human language), the team created a reliable system that can automatically read and pull useful data from thousands of articles. This innovative method not only helps us learn more about the variety of life on Earth, but also supports efforts to solve environmental challenges by making it easier to access important biological information.

Illustration depicting species literature feeding data on arthropod traits into a database, linking researchers and the community.
Mining the literature to identify species, their traits, and associated values.

The challenge

Scientific literature contains vast amounts of essential data about species—like what arthropods eat, where they live, and how big they are. However, this information is often trapped in hard-to-access files and old publications, making large-scale analysis almost impossible. So how can we convert these pages into usable data?

The goal

The team set out to develop an automatic text‑mining system using Natural Language Processing (NLP) and machine learning to scan thousands of biology papers and extract structured information about insects and other arthropods to build a database linking species names with traits like “leg length” or “forest habitat” or “predator”.

How it works in practice

  1. Collect curated vocabularies of terms to be searched for in the texts:
  • ~1 million species names from the Catalogue of Life
  • 390 traits, categorised into feeding ecology, habitat, and morphology 
  1. Create “Gold‑standard” data needed to train language models:
  • Experts manually annotated 25 papers—labelling species, traits, values, and their links—to use as a training benchmark
  1. Train NLP models so they “learn” which are the terms of interest:
  • Named‑Entity Recognition using BioBERT for identifying species, trait, and value words or phrases in the texts
  • Relation Extraction using LUKE to link the words/phrases e.g. “this species has this trait” and “this trait has this value” 
  1. Automated extraction of words/phrases and their links:
  • Processed 2,000 open‑access papers from PubMed Central
  • Identified ~656,000 entities (species, traits, values) and ~339,000 links between them 
  1. Publish results in an open searchable online resource:
  • Developed ArTraDB, an interactive web database where users can search, view, and visualise species‑trait pairs and full species‑trait‑value triples
Text-mining is a conceptually and computationally challenging task.

What is needed for the next steps

  • Annotation complexity: Even experts struggled to agree on boundaries and precise relationships, underscoring the need for clearer guidelines and more training examples to improve the performance of the models
  • Gaps in the vocabularies of terms: Many were unrecognised due to missing synonyms, outdated species names, and variations in phrasing. Expanding vocabularies will help improve the ability to find the species, traits, and values
  • Community curation: Planned features in ArTraDB will allow scientists and citizen curators to improve annotations, helping retrain and refine the models over time

How it impacts science

  • Speeds up research: Scientists can find species‑trait data quickly and accurately, boosting studies in ecology, evolution, and biodiversity
  • Scale and scope: This semi‑automated method can eventually be extended well beyond arthropods to other species
  • Supports global biodiversity efforts: Enables creation of large, quantitative trait datasets essential for monitoring ecosystem changes, climate impact, and conservation strategies
Illustration of a butterfly with icons and arrows outlining key biological data: barcode, genome, distribution, nutrition, habitat, and more.
A long-term vision to connect species with knowledge about their biology.

The outcomes

This innovative work demonstrates how combining text mining, expert curation, and interactive databases can unlock centuries of biological research. It lays a scalable foundation for building robust, open-access trait databases—empowering both scientists and the public to explore the living world in unprecedented ways.

Research article:

Cornelius J, Detering H, Lithgow-Serrano O, Agosti D, Rinaldi F, Waterhouse R (2025) From literature to biodiversity data: mining arthropod organismal traits with machine learning. Biodiversity Data Journal 13: e153070. https://doi.org/10.3897/BDJ.13.e153070