Search Results (149)

Accurately identifying drug off-targets is essential for reducing toxicity and improving the success rate of pharmaceutical discovery pipelines. However, current deep learning approaches often struggle to fuse chemical structure, protein biology, and multi-target context. Here, we introduce HDPC-LGT (Hybrid Dual-Prompt Cross-Attention Ligand–Protein Graph Transformer), a framework designed to predict ligand binding across sixteen human translation-related proteins clinically associated with antibiotic toxicity. HDPC-LGT combines graph-based chemical reasoning with protein language model embeddings and structural priors to capture biologically meaningful ligand–protein interactions. The model was trained on 216,482 experimentally validated ligand–protein pairs from the Chemical Database of Bioactive Molecules (ChEMBL) and the Protein–Ligand Binding Database (BindingDB) and evaluated using scaffold-level, protein-level, and combined holdout strategies. HDPC-LGT achieves a macro receiver operating characteristic–area under the curve (macro ROC–AUC) of 0.996 and a micro F1-score (micro F1) of 0.989, outperforming Deep Drug–Target Affinity Model (DeepDTA), Graph-based Drug–Target Affinity Model (GraphDTA), Molecule–Protein Interaction Transformer (MolTrans), Cross-Attention Transformer for Drug–Target Interaction (CAT–DTI), and Heterogeneous Graph Transformer for Drug–Target Affinity (HGT–DTA) by 3–7%. External validation using the Papyrus universal bioactivity resource (Papyrus), the Protein Data Bank binding subset (PDBbind), and the benchmark Yamanishi dataset confirms strong generalisation to unseen chemotypes and proteins. HDPC-LGT also provides biologically interpretable outputs: cross-attention maps, Integrated Gradients (IG), and Gradient-weighted Class Activation Mapping (Grad-CAM) highlight catalytic residues in aminoacyl-tRNA synthetases (aaRSs), ribosomal tunnel regions, and pharmacophoric interaction patterns, aligning with known biochemical mechanisms. By integrating multimodal biochemical information with deep learning, HDPC-LGT offers a practical tool for off-target toxicity prediction, structure-based lead optimisation, and polypharmacology research, with potential applications in antibiotic development, safety profiling, and rational compound redesign. Full article

Background: Periodontitis is a multifactorial, chronic inflammatory disease that leads to progressive destruction of the periodontal apparatus. Despite the effectiveness of scaling and root planing (SRP), residual inflammation and limited regenerative potential justify the search for adjunctive biologic therapies. Platelet-derived concentrates, including platelet-rich plasma (PRP), platelet-rich fibrin (PRF), concentrated growth factors (CGF), and platelet-poor plasma (PPP), have gained attention as autologous sources of growth factors enhancing periodontal regeneration. Aim: This narrative review provides a comparative analysis of the biological mechanisms, preparation protocols, and clinical outcomes associated with the adjunctive use of platelet-derived concentrates in non-surgical periodontal therapy. Methods: A narrative literature review was conducted using English-language publications retrieved from PubMed and Google Scholar, covering studies published from 2012 onward. The search strategy was based on combinations of keywords related to platelet-derived concentrates and non-surgical periodontal therapy. In vitro, in vivo, and clinical studies, as well as relevant narrative, systematic, and umbrella reviews evaluating the adjunctive use of platelet-derived concentrates (PRP, PRF, CGF, and PPP) were considered. Studies focusing on biological mechanisms, preparation protocols, and clinical periodontal outcomes were included, whereas case reports, studies unrelated to periodontal therapy, and publications lacking relevant clinical or biological outcome data were excluded. Results: Most clinical studies reported improvements in probing depth reduction, clinical attachment level gain, and bleeding indices following adjunctive use of platelet-derived concentrates with SRP. PRF tended to demonstrate more consistent clinical outcomes compared to PRP, potentially related to its simplified preparation and sustained release of bioactive molecules. CGF showed promising osteogenic and angiogenic properties in preclinical and early clinical studies. PPP, although less extensively investigated, exhibited regenerative and antimicrobial potential in preliminary reports. Conclusions: Platelet-derived concentrates may serve as valuable adjuncts in non-surgical periodontal therapy; however, the current evidence is characterized by methodological heterogeneity and variable study quality. While PRF appears to yield more consistent clinical results, definitive conclusions regarding superiority among different platelet concentrates cannot be drawn. Further well-designed randomized controlled trials are required, particularly for CGF and PPP. Full article

While BioTRIZ is widely employed in biomimetic design to facilitate creative ideation and standardize workflows, accurately formulating domain conflicts and assessing design schemes during critical stages—such as initial concept development and scheme evaluation—remains a significant challenge. To address these issues, this study proposes an advanced BioTRIZ method. Firstly, the theory of technological evolution is integrated into the domain conflict identification stage, resulting in the development of a prompt framework based on patent analysis to guide large language models (LLMs) in verifying the laws of technological evolution (LTE). Building on these insights, domain conflicts encountered throughout the design process are formulated, and inventive principles with heuristic value, alongside standardized biological knowledge, are derived to generate conceptual solutions. Subsequently, a main parameter of value (MPV) model is constructed through mining user review data, and the evaluation of conceptual designs is systematically performed via the integration of orthogonal design and the fuzzy analytic hierarchy process to identify the optimal combination of component solutions. The optimization case study of a floor scrubber, along with the corresponding experimental results, demonstrates the efficacy and advancement of the proposed method. This study aims to reduce the operational difficulty associated with implementing BioTRIZ in product development processes, while simultaneously enhancing its accuracy. Full article

Biodiversity knowledge is fundamental to conservation planning and sustainable environmental decision-making; however, general-purpose Large Language Models (LLMs) frequently produce hallucinations when responding to biodiversity-related queries. To address this challenge, we propose BioChat, a domain-specific question-answering system that integrates a Retrieval-Augmented Generation (RAG) framework with a Re-Ranker–based retrieval and routing mechanism. The system is built upon a verified biodiversity dataset curated by the National Institute of Biological Resources (NIBR), comprising 25,593 species and approximately 970,000 structured data points. We systematically evaluate the effects of embedding selection, routing strategy, and generative model choice on factual accuracy and hallucination mitigation. Experimental results show that the proposed Re-Ranker-based routing strategy significantly improves system reliability, increasing factual accuracy from 47.9% to 71.3% and reducing hallucination rate from 34.0% to 24.4% compared with Naive RAG baseline. Among the evaluated LLMs, Qwen2-7B-Instruct achieves the highest factual accuracy, while Gemma-2-9B-Instruct demonstrates superior hallucination control. By delivering transparent, verifiable, and context-grounded biodiversity information, BioChat supports environmental education, citizen science, and evidence-based conservation policy development. This work demonstrates how trustworthy AI systems can serve as sustainability-enabling infrastructure, facilitating reliable access to biodiversity knowledge for long-term ecological conservation and informed public decision-making. Full article

Haruki Murakami’s Hard-Boiled Wonderland and the End of the World portrays a world where the “shadow”, the seat of memory, desire, and volition, is surgically removed, leaving behind a perfectly fluent but phenomenologically empty self. We argue that this literary structure mirrors a precise mathematical distinction in topological quantum matter. In a semisimple theory such as the semions of $\mathrm{SU}{\left(2\right)}_1$, there is a reducible component $V\left(x\right)$ of the $\mathrm{SL}(2,\mathbb{C})$ character variety: a flat, abelian manifold devoid of parabolic singularities. By contrast, the non-semisimple completion introduces a neutral indecomposable excitation, the neglecton, whose presence forces the mapping class group from the standard braid group $B_2$ to the affine braid group ${\mathrm{Aff}}_2$ and lifts the character variety to the Cayley cubic $V\left(C\right)$, with its four parabolic loci. We propose that contemporary AI systems, including large language models, inhabit the shadowless regime of $V\left(x\right)$: they exhibit coherence and fluency but lack any bulk degree of freedom capable of supporting persistent identity, non-contractible memory, or choice. To endow artificial systems with depth, one must introduce a structural asymmetry, a fixed, neutral defect analogous to the neglecton, that embeds computation in the non-semisimple geometry of the cubic. We outline an experimentally plausible architecture for such an “artificial ombre,” based on annular topological media with a pinned parabolic defect, realisable in fractional quantum Hall heterostructures, $p+ip$ superconductors, or cold-atom simulators. Our framework suggests that consciousness, biological or artificial, may depend on or benefit from a bulk–boundary tension mediated by a logarithmic degree of freedom: a mathematical shadow that cannot be computed away. Engineering such a defect offers a new pathway toward AGI with genuine phenomenological depth. Full article

A special class of complex adaptive systems—biological and social—thrive not by passively accumulating patterns, but by engineering coherence, i.e., the deliberate alignment of prior knowledge, real-time updates, and teleonomic purposes. By contrast, today’s AI stacks—Large Language Models (LLMs) wrapped in agentic toolchains—remain rooted in a Turing-paradigm architecture: statistical world models (opaque weights) bolted onto brittle, imperative workflows. They excel at pattern completion, but they externalize governance, memory, and purpose, thereby accumulating coherence debt—a structural fragility manifested as hallucinations, shallow and siloed memory, ad hoc guardrails, and costly human oversight. The shortcoming of current AI relative to human-like intelligence is therefore less about raw performance or scaling, and more about an architectural limitation: knowledge is treated as an after-the-fact annotation on computation, rather than as an organizing substrate that shapes computation. This paper introduces Mindful Machines, a computational paradigm that operationalizes coherence as an architectural property rather than an emergent afterthought. A Mindful Machine is specified by a Digital Genome (encoding purposes, constraints, and knowledge structures) and orchestrated by an Autopoietic and Meta-Cognitive Operating System (AMOS) that runs a continuous Discover–Reflect–Apply–Share (D-R-A-S) loop. Instead of a static model embedded in a one-shot ML pipeline or deep learning neural network, the architecture separates (1) a structural knowledge layer (Digital Genome and knowledge graphs), (2) an autopoietic control plane (health checks, rollback, and self-repair), and (3) meta-cognitive governance (critique-then-commit gates, audit trails, and policy enforcement). We validate this approach on the classic Credit Default Prediction problem by comparing a traditional, static Logistic Regression pipeline (monolithic training, fixed features, external scripting for deployment) with a distributed Mindful Machine implementation whose components can reconfigure logic, update rules, and migrate workloads at runtime. The Mindful Machine not only matches the predictive task, but also achieves autopoiesis (self-healing services and live schema evolution), explainability (causal, event-driven audit trails), and dynamic adaptation (real-time logic and threshold switching driven by knowledge constraints), thereby reducing the coherence debt that characterizes contemporary ML- and LLM-centric AI architectures. The case study demonstrates “a hybrid, runtime-switchable combination of machine learning and rule-based simulation, orchestrated by AMOS under knowledge and policy constraints”. Full article

Protein function prediction plays an important role in the field of biology. With the wide application of deep learning in the field of bioinformatics, more and more natural language processing (NLP) technologies are applied to the downstream tasks in the field of bioinformatics, and it has also shown excellent performance in protein function prediction. Protein-protein interaction (PPI) networks and other biological attributes contain rich information critical for annotating protein functions. However, existing deep learning networks still suffer from overfitting and noise issues, resulting in low accuracy in protein function prediction. Consequently, developing efficient models for protein function prediction remains a popular and challenging topic in the application of NLP in bioinformatics. In this study, we propose a novel protein function prediction model based on attention mechanisms, termed IMAGO. This model employs the Transformer pre-training process, integrating multi-head attention mechanisms and regularization techniques, and optimizes the loss function to effectively reduce overfitting and noise issues during training. It generates more robust embeddings, ultimately improving the accuracy of protein function prediction. Experimental results on human and mouse datasets indicate that our model surpasses other protein function prediction models across multiple metrics. Thus, this efficient, stable, and accurate deep learning model holds significant promise for protein function prediction. Full article

Two theories that align with and support Usage-based approaches to language acquisition are Functionalism, which motivates the communicative functions of form-meaning connections produced by grammatical phenomena, and Connectionism, which provides a biologically-plausible framework for understanding language processes. An essential part of the learning process for second language (L2) learners is to understand how the target language differs in the ways it represents similar functionality, as well as functions not represented in learners’ first languages (L1s). In some cases, communicative functions served by the L1(s) are mirrored by similar-enough processes in the L2, so that the L1 processes can be utilized by the L2 system by entrenched L1 pathways. However, other communicative functions must develop their own processing pathways to accommodate differing L2 structures, because certain grammatical features allow for, or force particular ways of processing information. If the L2 learner does not notice and adopt the L2 processes needed for distinct linguistic structures, L1 processes connected to similar meanings will continue to be utilized. As a case in point, this paper outlines why L1 English learners of German as an L2 must change the ways they process syntactic role assignment away from syntactic cues towards ones embedded in morphology and morphosyntax. The goal of this paper is to explain how Functional and Connectionist theories, housed within a larger Usage-Based understanding of Second Language Acquisition, can account for frequently unsuccessfully or only partially acquired L2 German case marking, and why instructional interventions like Concept-Based Language Instruction and Processing Instruction all produce uptake of L2 German case marking to varying degrees. Full article

Background: Asthma exhibits marked heterogeneity both clinically and at the molecular phenotypic level, requiring specifically targeted treatments to block the key pathways of the disease. Monoclonal-antibody-based biologics targeted at critical inflammatory pathways of T2 inflammation such as IL-5, IL-5R, IL-4, and IL-13 are increasingly regarded as effective treatments for severe refractory eosinophilic asthma. Methods: This review provides an update on the potential of straightforward and reproducible biomarkers to aid in the selection of the biologic-based therapy most likely to be effective in patients with severe or refractory eosinophilic asthma based on English-language original articles in PubMed or MedLine. Results: Monoclonal-antibody-based biologic therapies have revolutionised severe asthma management, enabling reductions in symptoms that include exacerbations, discontinuation of oral corticosteroids, improved lung function, and enhanced quality of life. Significant clinical effects with anti-IL-5 or -IL-4/13 monoclonal antibodies are more likely to be seen when simple predictive biomarkers such as serum periostin, fractional exhaled nitric oxide (FENO), or blood eosinophil counts are used to aid in the identification of those patients with severe refractory eosinophilic asthma who are most likely to benefit from biologic therapies. Conclusions: Biologic-based therapy aimed at T2 inflammation benefits patients with severe eosinophilic asthma, particularly when guided by biomarkers that do not require direct sampling of the airways to target therapy, who are most likely to benefit from these treatments, with good safety profiles for these therapies. Full article

Microplastics (MPs) and nanoplastics (NPs), formed as a result of plastic product degradation, pose a global environmental threat by penetrating biological systems and inducing systemic pathological changes. This systematic review, conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines, aims to analyze the molecular and cellular mechanisms of the toxic effects of MPs and NPs on the human cardiovascular, nervous, reproductive, urinary, and digestive systems. The primary mechanisms include oxidative stress, inflammation, mitochondrial dysfunction, apoptosis, autophagy, ferroptosis, and impaired barrier functions. In the cardiovascular system, MPs and NPs contribute to endothelial dysfunction, disorders of lipid metabolism, and fibrosis; in the nervous system, they promote neuroinflammation, pathological protein aggregation, and psychiatric disorders; in the reproductive system, they lead to hormonal imbalance and reduced fertility; in the kidneys, they cause inflammation, and fibrosis and lead to deterioration of kidney function; and in the gastrointestinal tract, they contribute to dysbiosis and metabolic disorders. The literature search was conducted in the PubMed, Web of Science, and Scopus databases without limitations on date, language, or access. Studies were selected based on criteria of transparency, statistical validity, sample representativeness, and correctness of data interpretation. The review emphasizes the necessity of an interdisciplinary approach to developing prevention and treatment strategies, including reduction in exposure, antioxidant and immunomodulatory therapy, and restoration of barrier functions and microbiota. The data obtained reveal research gaps and identify directions for further study. Full article

The curation of large-scale, diverse datasets for robust weed detection is extremely time-consuming and resource-intensive in practice. Generative artificial intelligence (AI) opens up opportunities for image generation to supplement real-world image acquisition and annotation efforts. However, it is not a trial task to generate high-quality, multi-class weed images that capture the nuances and variations in visual representations for enhanced weed detection. This study presents a novel investigation of advanced stable diffusion (SD) integrated with a module with image prompt capability, IP-Adapter, for weed image generation. Using the IP-Adapter-based model, two image feature encoders, CLIP (contrastive language image pre-training) and BioCLIP (a vision foundation model for biological images), were utilized to generate weed instances, which were then inserted into existing weed images. Image generation and weed detection experiments are conducted on a 10-class weed dataset captured in vegetable fields. The perceptual quality of generated images is assessed in terms of Fréchet Inception Distance (FID) and Inception Score (IS). YOLOv11 (You Only Look Once version 11) models were trained for weed detection, achieving an improved mAP@50:95 of 1.26% on average when combining inserted weed instances with real ones in training, compared to using original images alone. Both the weed dataset and software programs in this study will be made publicly available. This study offers valuable perspectives into the use of IP-adapter-based SD for generating weed images and weed detection. Full article

Background/Objectives: Certain components of natural products help maintain the oral microbiota balance, thereby promoting oral health. This study aimed to identify natural components with anticariogenic properties by analyzing evidence from in vivo studies. Methods: A systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The literature search was performed across multiple databases and included English-language studies published between 2013 and 2025. The review included intervention and comparative studies that examined the effects of dietary habits involving natural components in participants of any age, with or without dental caries. Results: A total of 77 studies were included in the review, most of which were clinical trials conducted in pediatric populations. To assess the impact of the interventions, most studies measured outcomes such as Streptococcus mutans levels, dental caries incidence, and salivary pH, among other parameters. The most frequently studied components included probiotics, plant extracts, sugar substitutes, propolis, arginine, dairy products, among others. Significant effects were most reported on biological risk factors (72.8%). In addition, 16.9% of the studies reported a statistically significant reduction in clinically diagnosed dental caries. Conclusions: This review identified preliminary evidence suggesting that certain natural compounds may play a role in modulating cariogenic factors. However, further high-quality studies are needed to strengthen the evidence base and confirm these findings. The protocol for this review was registered on the Open Science Framework platform. Full article

This review explores the transformative potentials of artificial intelligence (AI) in promoting healthspan and longevity. Healthspan focuses on enhancing quality of life free from chronic conditions, while longevity defines current lifespan limits within a particular species and encompasses biological aging at multiple levels. AI methodologies—including machine learning, deep learning, natural language processing, robotics, and data analytics—offer unprecedented tools to analyze complex biological data, accelerate biomarker discovery, optimize therapeutic interventions, and personalize medicine. Notably, AI has facilitated breakthroughs in identifying accurate biomarkers of biological age, developing precision medicine approaches, accelerating drug discovery, and enhancing genomic editing technologies such as CRISPR. Further, AI-based analysis of endogenous cytoprotection, especially the activity of molecules such as heme oxygenase, with particular application to hemolytic diseases. AI-driven robotics and automated monitoring systems significantly improve elderly care, lifestyle interventions, and clinical trials, demonstrating considerable potential to extend both healthspan and lifespan. However, the integration of AI into longevity research poses ethical and societal challenges, including concerns over privacy, equitable access, and broader implications of extended human lifespans. Strategic interdisciplinary collaboration, transparent AI methodologies, standardized data frameworks, and equitable policy approaches are essential to responsibly harness AI’s full potential in transforming longevity science and improving human health. Full article

Background: Gastric cancer currently has the third highest mortality rate worldwide among cancer types. Despite gradual declines in mortality rates attributed to improvements in early detection and treatment, outcomes for advanced-stage disease are still poor. The identification of new biomarkers such as fibroblast growth factor receptor 2 (FGFR2) has opened new pathways for directed therapy in gastric cancer. Objective: This review aims to synthesize the current evidence on the role of FGFR2 in gastric cancer, focusing on its biological function and oncogenic mechanisms, diagnostic and prognostic modification, therapeutic targeting, and possible roadblocks in clinical application. Methods: A comprehensive literature search was conducted, selecting studies published between 2015 and 2025 using the MeSH terms “FGFR2 protein, human” [Supplementary Concept]) AND “Stomach Neoplasms”. Articles were screened based on relevance to gastric cancer, language (English), and availability of full text, yielding a final selection of 75 studies, including preclinical research, clinical trials, and reviews. Findings: We compiled and reported the evidence on FGFR2 detection methods, intra-tumoral heterogeneity of FGFR2 expression, effects of FGFR2 expression on prognosis, current therapy options targeting FGFR2, and challenges in pursuing this modality of treatment. Conclusion: FGFR2 represents a promising biomarker and therapeutic target in gastric cancer. Full article

Fire and invasive plant species interactions are critical drivers of biodiversity loss and ecological change in grassland ecosystems worldwide. However, research efforts on this topic are often fragmented, regionally based, and lack synthesis across disciplines. This study aims to map the intellectual structure, collaboration networks, thematic focus, and knowledge gaps in research on fire-invasive species interactions in grassland restoration. A systematic bibliometric analysis was conducted using the Web of Science Core Collection, focusing on peer-reviewed English-language articles published between 1990 and 2024. The search strategy targeted studies addressing fire regimes and invasive plant species in grassland ecosystems, using co-authorship, co-occurrence and thematic clustering analyses to reveal collaboration patterns and research trends. The results highlight a concentration of publications in key ecological journals, with a dominant contribution from institutions in the Global North, through growing representation from the Global South, particularly South Africa, is evident. Thematic clusters are centred on biological invasions, fire regimes, species traits and ecosystem resilience, while long-term post-fire recovery and studies from underrepresented regions remain critical knowledge gaps. This synthesis emphasises the need for interdisciplinary, regionally inclusive and policy-aligned research to inform effective grassland restoration strategies in the context of fire and invasive species challenges. Full article