Search

DeepLoc 2.0 is a popular web server for the prediction of protein subcellular localization and sorting signals. Here, we introduce DeepLoc 2.1, which additionally classifies the input proteins into the membrane protein types Transmembrane, Peripheral, Lipid-anchored and Soluble. Leveraging pre-trained transformer-based protein language models, the server utilizes a three-stage architecture for sequence-based, multi-label predictions. Comparative evaluations with other established tools on a test set of 4933 eukaryotic protein sequences, constructed following stringent homology partitioning, demonstrate state-of-the-art performance. Notably, DeepLoc 2.1 outperforms existing models, with the larger ProtT5 model exhibiting a marginal advantage over the ESM-1B model. The web server is available at https://services.healthtech.dtu.dk/services/DeepLoc-2.1.

When splitting biological sequence data for the development and testing of predictive models, it is necessary to avoid too-closely related pairs of sequences ending up in different partitions. If this is ignored, performance of prediction methods will tend to be overestimated. Several algorithms have been proposed for homology reduction, where sequences are removed until no too-closely related pairs remain. We present GraphPart, an algorithm for homology partitioning that divides the data such that closely related sequences always end up in the same partition, while keeping as many sequences as possible in the dataset. Evaluation of GraphPart on Protein, DNA and RNA datasets shows that it is capable of retaining a larger number of sequences per dataset, while providing homology separation on a par with reduction approaches.

When splitting biological sequence data for the development and testing of predictive models, it is necessary to avoid too-closely related pairs of sequences ending up in different partitions. If this is ignored, performance of prediction methods will tend to be overestimated. Several algorithms have been proposed for homology reduction, where sequences are removed until no too-closely related pairs remain. We present GraphPart, an algorithm for homology partitioning that divides the data such that closely related sequences always end up in the same partition, while keeping as many sequences as possible in the dataset. Evaluation of GraphPart on Protein, DNA and RNA datasets shows that it is capable of retaining a larger number of sequences per dataset, while providing homology separation on a par with reduction approaches., When splitting biological sequence data for the development and testing of predictive models, it is necessary to avoid too-closely related pairs of sequences ending up in different partitions. If this is ignored, performance of prediction methods will tend to be overestimated. Several algorithms have been proposed for homology reduction, where sequences are removed until no too-closely related pairs remain. We present GraphPart, an algorithm for homology partitioning that divides the data such that closely related sequences always end up in the same partition, while keeping as many sequences as possible in the dataset. Evaluation of GraphPart on Protein, DNA and RNA datasets shows that it is capable of retaining a larger number of sequences per dataset, while providing homology separation on a par with reduction approaches.

Little information is currently available on the occurrence and molecular diversity of the enteric protozoan parasites Cryptosporidium spp., Giardia duodenalis, and Balantioides coli in wild ungulates and the role of these host species as potential sources of environmental contamination and consequent human infections. The presence of these three pathogens was investigated in eight wild ungulate species present in Spain (genera Ammotragus, Capra, Capreolus, Cervus, Dama, Ovis, Rupicapra, and Sus) by molecular methods. Faecal samples were retrospectively collected from free-ranging (n = 1058) and farmed (n = 324) wild ungulates from the five Spanish bioregions. Overall infection rates were 3.0% (42/1382; 95% CI: 2.1–3.9%) for Cryptosporidium spp., 5.4% (74/1382; 95% CI: 4.2–6.5%) for G. duodenalis, and 0.7% (9/1382; 95% CI: 0.3–1.2%) for B. coli. Cryptosporidium infection was detected in roe deer (7.5%), wild boar (7.0%) and red deer (1.5%), and G. duodenalis in southern chamois (12.9%), mouflon (10.0%), Iberian wild goat (9.0%), roe deer (7.5%), wild boar (5.6%), fallow deer (5.2%) and red deer (3.8%). Balantioides coli was only detected in wild boar (2.5%, 9/359). Sequence analyses revealed the presence of six distinct Cryptosporidium species: C. ryanae in red deer, roe deer, and wild boar; C. parvum in red deer and wild boar; C. ubiquitum in roe deer; C. scrofarum in wild boar; C. canis in roe deer; and C. suis in red deer. Zoonotic assemblages A and B were detected in wild boar and red deer, respectively. Ungulateadapted assemblage E was identified in mouflon, red deer, and southern chamois. Attempts to genotype samples positive for B. coli failed. Sporadic infections by canine- or swine-adapted species may be indicative of potential cross-species transmission, although spurious infections cannot be ruled out. Molecular evidence gathered is consistent with parasite mild infections and limited environmental contamination with (oo)cysts. Free-ranging wild ungulate

Dolphins are marine mammals that often live in coastal habitats. Common causes of severe skeletal disorders among wild dolphins are congenital vertebral anomalities, collisions with sea vessels, trauma, hunting-related injury, infectious diseases, environmental pollution, and tumors. A free-ranging male, 3-year-old common dolphin (Delphinus delphis) was found dead in the coast of Asturias in northern Spain. Postmortem examination revealed lordosis in the caudal vertebral column, while X-ray imaging and computer tomography showed well-organized palisade-like periosteal proliferation, appearing as florid-like accretions, along the spinous apophysis of 26 lumbar-caudal vertebrae. The transverse apophysis was affected on only a few caudal vertebrae. The cortical layer remained intact. Histology of vertebra tissue showed periosteal proliferation of cancellous bone. The animal was diagnosed with hypertrophic osteopathy. The lungs showed diffuse parasitic granulomatous bronchointerstitial pneumonia caused by Halocercus delphini, consolidation of the pulmonary tissue, congestion, and alveolar edema. The animal was also afflicted by parasitic granulomatous gastritis caused by Anisakis simplex sensu lato and tattoo skin disease. The dolphin suffered from hypertrophic osteopathy associated with pulmonary Halocercus delphini infestation. This syndrome, known as hypertrophic pulmonary osteopathy, has been described in diverse terrestrial mammals, including domestic animals, wildlife and humans, but not in dolphins. This case reports the first description of hypertrophic osteopathy associated to a pulmonary disorder in dolphin, and it provides insights into factors that can induce column malformation in dolphins, suggesting the importance of taking thoracic lesions into account during differential diagnosis.

Microsporidia comprises a diverse group of obligate, intracellular, and spore-forming parasites that infect a wide range of animals. Among them, Enterocytozoon bieneusi is the most frequently reported species in humans and other mammals and birds. Data on the epidemiology of E. bieneusi in wildlife are limited. Hence, E. bieneusi was investigated in eight wild ungulate species present in Spain (genera Ammotragus, Capra, Capreolus, Cervus, Dama, Ovis, Rupicapra, and Sus) by molecular methods. Faecal samples were collected from free-ranging (n = 1058) and farmed (n = 324) wild ungulates from five Spanish bioregions. The parasite was detected only in red deer (10.4%, 68/653) and wild boar (0.8%, 3/359). Enterocytozoon bieneusi infections were more common in farmed (19.4%, 63/324) than in wild (1.5%, 5/329) red deer. A total of 11 genotypes were identified in red deer, eight known (BEB6, BEB17, EbCar2, HLJD-V, MWC_d1, S5, Type IV, and Wildboar3) and three novel (DeerSpEb1, DeerSpEb2, and DeerSpEb3) genotypes. Mixed genotype infections were detected in 15.9% of farmed red deer. Two genotypes were identified in wild boar, a known (Wildboar3) and a novel (WildboarSpEb1) genotypes. All genotypes identified belonged to E. bieneusi zoonotic Groups 1 and 2. This study provides the most comprehensive epidemiological study of E. bieneusi in Spanish ungulates to date, representing the first evidence of the parasite in wild red deer populations worldwide. Spanish wild boars and red deer are reservoir of zoonotic genotypes of E. bieneusi and might play an underestimated role in the transmission of this microsporidian species to humans and other animals.

The prediction of protein subcellular localization is of great relevance for proteomics research. Here, we propose an update to the popular tool DeepLoc with multi-localization prediction and improvements in both performance and interpretability. For training and validation, we curate eukaryotic and human multi-location protein datasets with stringent homology partitioning and enriched with sorting signal information compiled from the literature. We achieve state-of-the-art performance in DeepLoc 2.0 by using a pre-trained protein language model. It has the further advantage that it uses sequence input rather than relying on slower protein profiles. We provide two means of better interpretability: an attention output along the sequence and highly accurate prediction of nine different types of protein sorting signals. We find that the attention output correlates well with the position of sorting signals. The webserver is available at services.healthtech.dtu.dk/service.php?DeepLoc-2.0.

The prediction of protein subcellular localization is of great relevance for proteomics research. Here, we propose an update to the popular tool DeepLoc with multi-localization prediction and improvements in both performance and interpretability. For training and validation, we curate eukaryotic and human multi-location protein datasets with stringent homology partitioning and enriched with sorting signal information compiled from the literature. We achieve state-of-the-art performance in DeepLoc 2.0 by using a pre-trained protein language model. It has the further advantage that it uses sequence input rather than relying on slower protein profiles. We provide two means of better interpretability: an attention output along the sequence and highly accurate prediction of nine different types of protein sorting signals. We find that the attention output correlates well with the position of sorting signals. The webserver is available at services.healthtech.dtu.dk/service.php?DeepLoc-2.0.

Multi-host pathogens emerging and re-emerging at the wildlife–domestic animal interface affect wildlife management and conservation. This is the case of canine distemper virus (CDV), a paramyxovirus closely related to human measles virus and rinderpest virus of cattle. With an area of 10,603 km2, Asturias region in Atlantic Spain is a hotspot of carnivore diversity, which includes the largest Eurasian brown bear (Ursus arctos arctos) population and one of the largest wolf (Canis lupus) populations in south-western Europe. In 2020–2021, we recorded mortality due to distemper in four carnivore species including three mustelids (Eurasian badger Meles meles, European marten Martes martes and European polecat Mustela putorius) and one canid (red fox, Vulpes vulpes). Clinical signs and pathology were similar across species and consistent with the emergence of a highly pathogenic viral strain, with CDV antigen mainly located in the central nervous system, lungs, spleen and lymph nodes. A molecular study in eight wild carnivore species, also including the Iberian wolf, Eurasian brown bear, American mink (Neovison vison) and stone marten (Martes foina), revealed 19.51% (16/82) of positivity. Phylogenetic analysis demonstrated that CDV belonged to the previously described European lineage. A retrospective serosurvey (2008–2020) showed a high seroprevalence of CDV antibodies (43.4%) in 684 analyzed badgers, indicating a long-term though not stable viral circulation in this multi-host community. The possible triggers of the 2020–2021 outbreak and the implications for carnivore management and conservation are discussed.

The present work investigated the prevalence, spatial distribution, and temporal distribution of tuberculosis (TB) in free-ranging Eurasian badgers (Meles meles) and cattle in Asturias (Atlantic Spain) during a 13-year follow-up. The study objective was to assess the role of badgers as a TB reservoir for cattle and other sympatric wild species in the region. Between 2008 and 2020, 673 badgers (98 trapped and 575 killed in road traffic accidents) in Asturias were necropsied, and their tissue samples were cultured for the Mycobacterium tuberculosis complex (MTC) isolation. Serum samples were tested in an in-house indirect P22 ELISA to detect antibodies against the MTC. In parallel, data on MTC isolation and single intradermal tuberculin test results were extracted for cattle that were tested and culled as part of the Spanish National Program for the Eradication of Bovine TB. A total of 27/639 badgers (4.23%) were positive for MTC based on bacterial isolation, while 160/673 badgers (23.77%) were found to be positive with the P22 ELISA. The rate of seropositivity was higher among adult badgers than subadults. Badger TB status was spatially and temporally associated with cattle TB status. Our results cannot determine the direction of possible interspecies transmission, but they are consistent with the idea that the two hosts may exert infection pressure on each other. This study highlights the importance of the wildlife monitoring of infection and disease during epidemiological interventions in order to optimize outcomes., Ministerio de Ciencia e Innovación (MICINN)/FEDER, Depto. de Producción Animal, Fac. de Veterinaria, TRUE, pub

Provision of food and water is a widespread tool implemented around the world for the benefit of game and other wildlife, but factors affecting the use of food and water by non-target species are poorly known. We evaluated visits to feeders and water troughs by non-game species using camera-traps in two separate areas of Spain. Feeders and water troughs were either “protected” (when surrounded by more than 50% of shrubs/forest) or “open” (in the opposite case). A total of 18 948 photos from 5344 camera-trapping days depicted animals, and 75 species were identified. Feeders and water troughs were visited by target species (partridges and lagomorphs, 55.3% of visits) and non-target species (44.7% of visits). Among the latter, corvids were the most common (46.1% of visits), followed by rodents (26.8%), other birds (23.6%, mainly passerines), columbids (1.9%), and other species at minor percentages. The highest proportion of visiting days to feeders and water troughs was from corvids (0.173) followed by other-birds (0.109) and rodents (0.083); the lowest proportion was recorded for columbids (0.016). Use intensity and visit frequency of water troughs tripled that recorded in feeders, and visits to open feeders/troughs were approximately twice those to protected ones. In summary: feeders and water troughs targeting small game species are also used regularly by non-target ones; they should be set close to cover to optimize their use by non-target species that are not competitors of target species (though corvids may visit them); water availability should be prioritized where drought periods are expected.

Signal peptides (SPs) are short amino acid sequences in the amino terminus of many newly synthesized proteins that target proteins into, or across, membranes. Bioinformatic tools can predict SPs from amino acid sequences, but most cannot distinguish between various types of signal peptides. We present a deep neural network-based approach that improves SP prediction across all domains of life and distinguishes between three types of prokaryotic SPs.

The red deer (Cervus elaphus) is a widespread wild ungulate in Europe that has suffered strong anthropogenic impacts over their distribution during the last centuries, but also at the present time, due its economic importance as a game species. Here we focus on the evolutionary history of the red deer in Iberia, one of the three main southern refugial areas for temperate species in Europe, and addressed the hypothesis of a cryptic refugia at higher latitudes during the Last Glacial Maximum (LGM). A total of 911 individuals were sampled, genotyped for 34 microsatellites specifically developed for red deer and sequenced for a fragment of 670 bp of the mitochondrial (mtDNA) D-loop. The results were combined with published mtDNA sequences, and integrated with species distribution models and historical European paleo-distribution data, in order to further examine the alternative glacial refugial models and the influence of cryptic refugia on European postglacial colonization history. Clear genetic differentiation between Iberian and European contemporary populations was observed at nuclear and mtDNA levels, despite the mtDNA haplotypes central to the phylogenetic network are present across western Europe (including Iberia) suggesting a panmictic population in the past. Species distribution models, fossil records and genetic data support a timing of divergence between Iberian and European populations that overlap with the LGM. A notable population structure was also found within the Iberian Peninsula, although several populations displayed high levels of admixture as a consequence of recent red deer translocations. Five D-loop sub-lineages were found in Iberia that belong to the Western European mtDNA lineage, while there were four main clusters based on analysis of nuclear markers. Regarding glacial refugial models, our findings provide detailed support for the hypothesis that red deer may have persisted in cryptic northern refugia in western Europe during the LGM, most