Dyrosauridae are one of the few clades of large marine vertebrates that survived the Cretaceous/Palaeogene (K–Pg) extinction event. The early diversification of this clade, dated to the Cretaceous, is still poorly understood. We describe a new specimen of Dyrosauridae from the Late Cretaceous of Bentiaba, Angola, represented by a posterior portion of the skull, a left ectopterygoid fragment, and a tooth. Comparative morphological analysis reveals a unique combination of characters distinguishing it from known dyrosaurid taxa but due to the fragmentary nature of the specimen and the absence of key diagnostic features, we assign it to Hyposaurinae indet.. Phylogenetic relationships were tested through maximum parsimony and Bayesian inference, using the SFBD model with time-binning criteria for the K–Pg boundary. Both of our phylogenetic analyses recover this new specimen as Hyposaurinae, which is consistent with our comparative study. Evolutionary rates accelerated during the latest stages of the Cretaceous, although the fossil bias prevents the ability to appreciate their impact on cladogenetic events. We estimated diversification, fossilization, and extinction rates, and we did not find evidence of radiation after the K–Pg in this clade. Biogeographically, this discovery supports hypotheses of longirostrine crocodyliform dispersal in Africa leading to hyposaurine diversification during the Maastrichtian.

Polysaccharides in plant cell walls serve as a rich carbon and energy source, yet their structural complexity presents a barrier to efficient degradation. To address this, anaerobic microorganisms like R. flavefaciens have developed sophisticated multi-enzyme complexes known as cellulosomes, which enable the efficient breakdown of these recalcitrant polysaccharides. These complexes are assembled through high-affinity interactions between cohesin (Coh) modules in scaffoldin proteins and dockerin (Doc) modules in cellulosomal enzymes. R. flavefaciens FD-1 harbors one of the most intricate cellulosomes described to date, comprising over 200 Doc-containing proteins encoded in its genome. Despite substantial research on this cellulosome, the role of a group of truncated but functional dockerins, known as group-2 Docs, remains unclear. In this study, we present a detailed structural and binding analysis of a Coh-Doc complex involving the cohesin from the cell-anchoring scaffoldin ScaE and a group-2 Doc that bears only one of the two Ca+2-coordinating loops that characterise the canonical Docs. Our findings reveal a novel tripartite binding mechanism, in which the cohesin can simultaneously bind two distinct dockerin units in three alternative conformations. This discovery provides new insights into the modular versatility of the R. flavefaciens cellulosome and sheds light on the mechanisms that enhance its efficiency in polysaccharide degradation.

High-dimensional omics data often contain more variables than observations, which can lead to overfitting and negatively impact the results of classical data analysis methods. To address the issue, supervised variable selection methods are often used, incorporating penalty terms into the model. While effective for selecting task-specific variables, this approach may not preserve the overall dataset structure for multiple downstream analyses. This study aims to evaluate unsupervised variable selection approaches and introduce a novel tool that improves data interpretability while maintaining biological information.

The potential of STE(A)M has been widely recognised in recent years; however, challenges have also been identified in the implementation of this approach, making it important to deepen research into teachers’ practices. In this study, we conducted a systematic review focused on teaching practices within a STE(A)M approach, to understand what challenges and limitations teachers encounter, as well as the possibilities authors suggest to overcome them. The research question is: What are the challenges and potentialities of implementing the STE(A)M approach in teachers’ practices? This review analysed articles indexed in the Web of Science and Scopus databases over the last eight years, following the PRISMA scheme. The 26 articles examined highlight challenges such as time management, lack of resources and funding, limited guidance, gaps in teacher training, difficulty integrating subject areas, issues implementing new technologies and obstacles imposed by school structures. In light of these, the following proposals were suggested: placing greater emphasis on STE(A)M in initial and continuing teacher education, increasing support from schools and government bodies, encouraging collaborative work between teachers from different areas, adopting innovative strategies and methodologies and offering a clearer conceptualisation of STE(A)M.

The Upper Jurassic beds of the Lusitanian Basin in central Portugal yield diverse dinosaurian fauna, dated to the Kimmeridgian–Tithonian interval. Saurischian dinosaurs are, overall, more abundant than their ornithischian counterparts, in terms of both specimens collected and species recognized. Iguanodontians are so far represented by the styracosternan Draconyx loureiroi, the dryosaurid Eousdryosaurus nanohallucis and the enigmatic dryomorphan Hesperonyx martinhotomasorum. Here we aim to highlight the diversity of this clade in the Late Jurassic of Portugal, presenting evidence for yet another species of ankylopollexian iguanodontian dinosaur, represented by the specimen SHN.JJS.015, which is housed at the Sociedade de História Natural, Torres Vedras. Detailed comparisons rule out attribution to previously known taxa, and phylogenetic analyses that include SHN.JJS.015 indicate early-diverging ankylopollexian affinities for this specimen. As there is no robust diagnosis, we do not erect a new formal species for it at this stage. Nevertheless, this specimen represents a previously unreported taxon that highlights greater diversity than previously estimated among the iguanodontians of the Late Jurassic and highlights the importance of Europe in diversification and dispersal events of this clade. A series of smaller, isolated femora from the same sub-basin as SHN.JJS.015 may represent the same taxon, presenting evidence of thriving communities of ankylopollexians during the Kimmeridgian–Tithonian interval in Portugal.

Incorporating technologies with Generative Artificial Intelligence (GenAI) into education requires a shift in teaching methodologies. However, little is known about how pre-service teachers perceive the relevance and challenges of this incorporation, particularly in mathematics education. This study investigates pre-service teachers’ interactions with GenAI, addressing the relevance and challenges of integrating it in mathematics teaching and discussing possible implications for their knowledge and professional development. Specifically, it intends to understand: How does the pre-service teachers’ interaction with GenAI during the design and implementation of teaching activities relate to their professional knowledge? And how does this relation impact the relevance they ascribe to GenAI? In this qualitative and interpretative study, involving seven pre-service mathematics teachers, we analyse the interplay between participants’ knowledge and use of a GenAI (in this case, ChatGPT), guided by KTMT – Knowledge for Teaching Mathematics with Technology model. The main conclusions revealed a landscape characterised by promise and challenge, where GenAI can be a valuable educational
tool when used to facilitate discussion and promote critical thinking, highlighting the relevance and development of KTMT. The ability to evaluate and reflect on AI-generated responses can promote professional development, preparing pre-service teachers for an increasing presence of technology in educational environments.

The teacher’s central role in technology integration and the challenges of that integration emphasise the need for a deeper understanding about the teacher’s knowledge required to teach with technology. Based on previous work and a systematic literature review, we identified three knowledge models often used: TPACK, KTMT and
PTK. The goal of this paper is to discuss the similarities and differences between these knowledge models and present a Global Model. This Global Model is not a new model. On the contrary, it is a model developed based on the existing models and intending to integrate in a single model the knowledge domains considered in the different existing models. The Global Model highlights the common domains considered and the common roots for the three models, but it also makes explicit the differences, mostly related to the understanding of the domains or even to the domains considered, and also to the way how the knowledge’s development is conceived.

ABSTRACT The first mammaliaforms emerged in the Late Triassic, but their exact origins remain unclear due to the scarcity of fossils from this period. One of the earliest diverging mammaliaform groups, the order Docodonta, became unusually ecomorphologically diverse compared with other early mammals, and this may be connected to the possession of complex molar cusp morphology. The specimen described here, found in the Rhætelv Formation of the Kap Stewart Group (Rhaetian–Sinemurian) of central East Greenland, provides novel information on docodontan origins and evolution, as well as key biogeographic insights into early mammal dispersal. Nujalikodon cassiopeiae gen. et sp. nov. is the first mammaliaform found in the Rhætelv Formation, and is likely to be Early Jurassic (Hettangian) in age. Comprising an incomplete dentary with a single preserved molar, it was visualized using micro-computed tomography; the molar bears similarities to the putative early docodontan Delsatia, and docodontan Dobunnodon. Phylogenetic analysis places Nujalikodon cassiopeiae as a basal member of Docodonta or a close sister taxon, making it one of the oldest definitive docodontans and pushing the origin of the group back to at least the Early Jurassic. It provides insights into the development of docodontan dental complexity, a key factor in their ecological diversification during the Middle to Late Jurassic. Its presence in Greenland supports the hypothesis that docodontans originated in the region now comprising Europe and Greenland before dispersing across the rest of Laurasia.

The teachers’ central role in technology integration and the challenges of that integration emphasize the need for a deeper understanding about the teachers’ knowledge required to teach with technology. Based on previous work and a systematic literature review, we identified three knowledge models often used: TPACK, KTMT and PTK. The goal of this paper is to discuss the similarities and differences between these knowledge models and present a Global Model. This Global Model is not a new model. On the contrary, it is a model developed based on the existing models and intending to integrate in a single model the knowledge domains considered in the different existing models. The Global Model highlights the common domains considered and the common roots for the three models, but it also makes explicit the differences, mostly related to the understanding of the domains or even to the domains considered, and also to the way how the knowledge’s development is conceived.

Accurate blast models are essential for disaster management and emergency preparedness. Semi-empirical methods, which rely on free-field assumptions, struggle to provide precise data for complex building shapes because they ignore shock wave reflections. This research numerically investigates the effects of an explosion on a large building of complex geometrical configuration using blastFoam, estimating the façade damage and assessing risks to occupants. To evaluate the importance of accurately modelling the building’s geometry, four levels of detail were considered. Additionally, simulations were also performed using the Load Blast Enhanced (LBE) method from LS-DYNA to compare the estimates of this faster semi-empirical approach with those obtained through CFD analyses. The findings reveal that simpler CFD models are adequate for façade analysis and injury assessment around the building but fall short for predicting injury distribution within interior spaces or between buildings. The LBE underestimates both structural damage and human injury levels, while detailed CFD highlights the importance of accounting for interior walls and windows to enhance blast pressure predictions.

In this work, fully patterned zinc tin oxide (ZTO) memristors are introduced using inkjet printing. By targeting a scalable, solution-based fabrication approach, highly stable devices with excellent reproducibility and minimal variability are achieved, using ZTO as the active layer, silver (Ag) as the top electrode, and molybdenum as the bottom electrode. The use of sustainable materials like ZTO enhances scalability and environmental compatibility, paving the way for next-generation, low-power neuromorphic computing. The devices successfully fulfill the fundamental criteria for in materia implementation of physical reservoir computing (PRC), including nonlinearity and fading memory property. The devices are successfully trained for classification tasks with MNIST handwritten dataset, achieving 89.4% accuracy and 86.5% by processing 4-bit and 5-bit input temporal sequences. The integration of printed memristors into hardware-based PRC architecture simplifies training complexity, making them particularly advantageous for energy-efficient, wearable AI systems.