Search Results (557)

Reversible protein phosphorylation is an important regulatory mechanism in cellular signalling and disease, regulated by the opposing actions of kinases and phosphatases. Modern computer methods predict kinase–substrate or phosphatase–substrate interactions in isolation and lack specificity for biological conditions, neglecting triadic regulation. We present SPINET-KSP, a multi-modal LLM–Graph foundation model engineered for the prediction of kinase–substrate–phosphatase (KSP) triads with contextual awareness. SPINET-KSP integrates high-confidence interactomes (SIGNOR, BioGRID, STRING), structural contacts obtained from AlphaFold3, ESM-3 sequence embeddings, and a 512-dimensional cell-state manifold with 1612 quantitative phosphoproteomic conditions. A heterogeneous KSP graph is examined utilising a cross-attention Graphormer with Reversible Triad Attention to mimic kinase–phosphatase antagonism. SPINET-KSP, pre-trained on 3.41 million validated phospho-sites utilising masked phosphorylation modelling and contrastive cell-state learning, achieves an AUROC of 0.852 for kinase-family classification (sensitivity 0.821, specificity 0.834, MCC 0.655) and a Pearson correlation coefficient of 0.712 for phospho-occupancy prediction. In distinct 2025 mass spectrometry datasets, it identifies 72% of acknowledged cancer-resistance triads within the top 10 rankings and uncovers 247 supplementary triads validated using orthogonal proteomics. SPINET-KSP is the first foundational model for simulating context-dependent reversible phosphorylation, enabling the targeting of dysregulated kinase-phosphatase pathways in diseases. Full article

High-resolution climate projections are essential for regional risk assessment; however, Earth System Models (ESMs) operate at scales far too coarse for local impacts. This review examines how machine learning (ML) downscaling can bridge this divide and addresses a key knowledge gap: how to achieve reliable, physically consistent downscaling under future climate change. This article synthesizes ML downscaling developments from 2010 to 2025, spanning early statistical methods to modern deep learning (e.g., convolutional neural networks (CNNs), generative adversarial networks (GANs), diffusion models, and transformers). The analysis introduces a new taxonomy of model families and frames the discussion around the “performance paradox”—the tendency for models with excellent historical skill to falter under non-stationary climate shifts. Our analysis finds that convolutional approaches efficiently capture spatial structure but tend to smooth out extremes, whereas generative models better reproduce high-intensity events at the cost of greater complexity. The study also highlights emerging solutions like physics-informed models and improved uncertainty quantification to tackle persistent issues of physical consistency and trust. Finally, the synthesis outlines a practical roadmap for operational ML downscaling, emphasizing standardized evaluation, out-of-distribution stress tests, and hybrid physics–ML approaches to bolster confidence in future projections. Full article

Drug–target affinity (DTA) prediction is one of the core components of drug discovery. Despite considerable advances in previous research, DTA tasks still face several limitations with insufficient multi-modal information of drugs, the inherent sequence length limitation of protein language models, and single attention mechanisms that fail to capture critical multi-scale features. To alleviate the above limitations, we developed a multi-granularity fusion model for drug–target binding affinity prediction, termed MGF-DTA. This model is composed of three fusion modules, specifically as follows. First, the model extracts deep semantic features of SMILES strings through ChemBERTa-2 and integrates them with molecular fingerprints by using gated fusion to enhance the multi-modal information of drugs. In addition, it employs a residual fusion mechanism to integrate the global embeddings from ESM-2 with the local features obtained by the k-mer and principal component analysis (PCA) method. Finally, a hierarchical attention mechanism is employed to extract multi-granularity features from both drug SMILES strings and protein sequences. Comparative analysis with other mainstream methods on the Davis, KIBA, and BindingDB datasets reveals that the MGF-DTA model exhibits outstanding performance advantages. Further, ablation studies confirm the effectiveness of the model components and case study illustrates its robust generalization capability. Full article

Spatial cognition refers to the mental processing, perception, and interpretation of spatial information. It is often operationalized through self-assessments like sense of direction and mental rotation ability or field-based real-world tasks like pointing to a specific building and wayfinding; however, the former and latter entail unclear ecological validity and high participant burdens, respectively. Since the advent of smartphones, this repertoire has been extended substantially through the use of sensors or apps. This study used a large longitudinal experience sampling method (ESM) in two different countries (Canada and Australia, N = 217) and analyzed spatial cognition both conventionally (i.e., sense of direction and speeded mental rotation test) and through new techniques like self-rated and objectively assessed daily Google Maps usage, movement patterns throughout the 14-day assessment phase (using H3 tiles for geolocation), and a Point North task. The Point North task objectively assessed deviation from the celestial direction, North, by using smartphone compass sensors. In both countries, spatial orientation was found to be associated only with the Point North task, while no significant associations were found for daily Google Maps usage (subjectively and objectively measured) and moving distance throughout the assessment phase. Although further validation is required, the Point North task shows promise as an objective, ecologically valid, and easily employable smartphone-based measure for assessing spatial cognition in real-world contexts. Full article

Saxitoxin (STX) is one of the most potent marine neurotoxins, produced by several species of freshwater cyanobacteria and marine dinoflagellates. Although omics-based approaches have advanced our understanding of STX biosynthesis in recent decades, the origin, regulation, and ecological drivers of STX in dinoflagellates remain poorly resolved. Specifically, dinoflagellate STX biosynthetic genes (sxt) are extremely fragmented, inconsistently expressed, and unevenly distributed between toxic and non-toxic taxa. Environmental studies further report inconsistent relationships between abiotic factors and STX production, suggesting regulation across multiple genomic, transcriptional, post-transcriptional, and epigenetic levels. These gaps prevent a comprehensive understanding of STX biosynthesis in dinoflagellates and limit the development of accurate predictive models for harmful algal blooms (HABs) and paralytic shellfish poisoning (PSP). Artificial intelligence (AI), including machine learning and deep learning, offers new opportunities in ecological pattern recognition, molecular annotation, and data-driven prediction. This review explores the current state of knowledge and persistent knowledge gaps in dinoflagellate STX research and proposes an AI-integrated multi-omics framework highlighting recommended models for sxt gene identification (e.g., DeepFRI, ProtTrans, ESM-2), evolutionary reconstruction (e.g., PhyloGAN, GNN, PhyloVAE, NeuralNJ), molecular regulation (e.g., MOFA+, LSTM, GRU, DeepMF), and toxin prediction (e.g., XGBoost, LightGBM, LSTM, ConvLSTM). By integrating AI with diverse biological datasets, this novel framework outlines how AI can advance fundamental understanding of STX biosynthesis and inform future applications in HAB monitoring, seafood safety, and PSP risk management in aquaculture and fisheries. Full article

The increasing penetration of variable renewable energy sources has intensified the need for ancillary services to maintain grid stability, and demand-side flexibility, particularly through distributed energy systems (DESs), is expected to play an important role. This study proposes a two-stage optimization framework for DESs under CO₂ constraints that enables gas engines and battery energy storage systems (BESS) to provide regulating power equivalent to Load Frequency Control (LFC). The framework consists of an Equipment Sizing Optimization Model (ESM) and an Equipment Operation Optimization Model (EOM), both formulated as mixed-integer linear programming (MILP) models. The ESM determines equipment capacities using simplified operational representations, where partial-load efficiencies are approximated through linear programming (LP)-based constraints. The EOM incorporates detailed operational characteristics, including start-up/shutdown states and partial-load efficiencies, to perform daily scheduling. Information obtained from the ESM, such as the CO₂ emissions, the equipment capacities, and the BESS state of charge, is passed to the EOM to maintain consistency. A case study shows that providing regulating power reduces total system cost and that CO₂ reduction constraints alter the equipment mix. These findings demonstrate that the proposed framework offers a practical and computationally efficient approach for designing and operating DESs under CO₂ constraints. Full article

Background: Machine learning models have been shown to be a time-saving and cost-effective tool for peptide-based drug discovery. In this regard, different graph learning-driven frameworks have been introduced to exploit graph representations derived from predicted peptide structures. Such graphs are always derived by applying a Euclidean distance threshold between amino acid pairs, despite the fact that there is no evidence other than intuitive reasoning that supports the Euclidean distance as the most suitable. Objective: In this work, we examined the use of different distance functions to derive graph representations from predicted peptide structures to train deep graph learning-based models to predict antiviral peptides. Methods: To this end, we first analyzed how differently the closeness of the amino acids is characterized by different distance functions. Then, we studied the similarity between the graphs derived with several distance functions, as well as between them and random graphs. Finally, we trained several models with the best graph representations and analyzed how different they are regarding their predictions. Comparisons regarding state-of-the-art models were also performed. Results and Conclusion: We demonstrated that only using Euclidean distance thresholds is not sufficient criterion to build graphs representing structural features of predicted peptide structures, since other distance functions enabled building dissimilar graphs codifying different chemical spaces, which were useful in the construction of better discriminative models. Full article

The pathophysiology of ST-elevated myocardial infarction (STEMI) extends beyond coronary artery occlusion to include microvascular and endothelial dysfunction, both of which critically influence outcomes. Endocan, a soluble dermatan sulfate proteoglycan secreted by endothelial cells, has emerged as a novel biomarker of endothelial activation and dysfunction. Recent studies suggest that elevated endocan levels may carry prognostic significance in patients with STEMI, particularly those undergoing percutaneous coronary intervention (PCI). A comprehensive search of PubMed, Cochrane Library, and Google Scholar was conducted to identify studies evaluating endocan as a prognostic biomarker in STEMI. Review articles, case reports, case series, and experimental studies were excluded. Seven clinical studies, comprising sample sizes ranging from 80 to 320 patients, met the inclusion criteria. Across these studies, endocan levels were analyzed in relation to established prognostic markers and clinical outcomes. Key findings demonstrated that higher endocan levels correlated with stress hyperglycemia (r = 0.21, p < 0.05), higher SYNTAX scores, and worse in-hospital outcomes. A cutoff value of 1.7 ng/mL predicted STEMI with 76.1% sensitivity and 73.6% specificity. Elevated endocan levels also showed positive correlations with the TIMI risk score, major adverse cardiovascular events (MACE), and were identified as independent predictors of incomplete ST-segment resolution (STR) (p = 0.044) and no-reflow phenomenon (NRP) (p < 0.001, OR = 2.39, 95% CI = 1.37–4.15). Collectively, the evidence indicates that endocan is strongly associated with endothelial dysfunction, MACE, NRP post-PCI, and impaired reperfusion. Moreover, traditional prognostic indices such as TIMI and SYNTAX scores appear to correlate with circulating endocan levels. However, variability in reported cutoff values across studies highlights the need for larger, multicenter trials with standardized endpoints to establish endocan’s diagnostic and prognostic utility in STEMI. Full article

Enzyme engineering drives innovation in biotechnology, medicine, and industry, yet conventional approaches remain limited by labour-intensive workflows, high costs, and narrow sequence diversity. Artificial intelligence (AI) is revolutionising this field by enabling rapid, precise, and data-driven enzyme design. Machine learning and deep learning models such as AlphaFold2, RoseTTAFold, ProGen, and ESM-2 accurately predict enzyme structure, stability, and catalytic function, facilitating rational mutagenesis and optimisation. Generative models, including ProteinGAN and variational autoencoders, enable de novo sequence creation with customised activity, while reinforcement learning enhances mutation selection and functional prediction. Hybrid AI–experimental workflows combine predictive modelling with high-throughput screening, accelerating discovery and reducing experimental demand. These strategies have led to the development of synthetic “synzymes” capable of catalysing non-natural reactions, broadening applications in pharmaceuticals, biofuels, and environmental remediation. The integration of AI-based retrosynthesis and pathway modelling further advances metabolic and process optimisation. Together, these innovations signify a shift from empirical, trial-and-error methods to predictive, computationally guided design. The novelty of this work lies in presenting a unified synthesis of emerging AI methodologies that collectively define the next generation of enzyme engineering, enabling the creation of sustainable, efficient, and functionally versatile biocatalysts. Full article

The ramifications of climate change, which are projected to lead to increased drought, desertification, and water scarcity, are expected to have a significant impact on the agricultural sector of Türkiye, particularly in the Mediterranean coastal regions. This study presents an extensive evaluation of potential agricultural drought conditions in southwestern Türkiye, using a high-resolution, convection-permitting (0.025°) modeling approach. We employ a single, physically consistent model chain, dynamically downscaling the CMIP6 MPI-ESM-HR Earth System Model with the COSMO-CLM regional climate model at a convection-permitting (CP) resolution (0.025°) under IPCC Shared Socioeconomic Pathways SSP3-7.0, reflecting a high-emission scenario with regional socioeconomic challenges. Southwestern Türkiye, situated at the intersection of the Mediterranean and continental climates, hosts rare climatic and ecological conditions that sustain a highly productive and diverse agricultural system. This region forms the backbone of Türkiye’s agricultural economy but is increasingly vulnerable to climate variability and fluctuations that threaten its agricultural stability and resilience. Our study employs a novel approach that utilizes multivariate assessment of agricultural drought in the Mediterranean Region by integrating precipitation, soil moisture, and temperature variables from 2.5 km resolution climate simulations. Agricultural drought conditions were evaluated using the Standardized Precipitation Index (SPI), the Standardized Soil Moisture Index (SSI), and the Standardized Temperature Index (STI), derived by normalizing respective climate variables from climate simulations spanning from 1995 to 2014 for the historical period, from 2040 to 2049 and from 2070 to 2079 for future projections. CP climate simulations (CPCSs) exhibit a modest warm and dry bias during all seasons but slightly wetter conditions during summer when compared with station observations. Correlations between indices indicate that soil moisture variations in the future will become more sensitive to changes in temperature rather than precipitation. Results from this specific model chain reveal that the probability of compound events where precipitation and soil moisture deficits coincide with anomalously high temperatures will rise for all threshold levels under the SSP3-7.0 scenario towards the end of the century. For the most severe conditions (|Z| > 1.2), the compound likelihood increases to about 3%, highlighting the enhanced occurrence of rare events in a changing climate. These findings, conditional on the model and scenario used, provide a high-resolution, physically grounded perspective on the potential intensification of agricultural drought regimes. Full article

Global glacier models (GGMs) are effective tools for assessing changes in water resources in mountainous regions and studying glacier degradation. Moreover, with the rapid development and increasing complexity of Earth System Models (ESMs), the incorporation of mountain glaciation parametrizations into ESMs is only a matter of time. GGMs, being computationally efficient and physically well-founded, provide a solid basis for such parametrizations. In this study, we present a new global glacier model, IGRICE. Its dynamic core is based on the Oerlemans minimal model, and surface mass balance (SMB) is explicitly simulated, accounting for orographic precipitation, radiation redistribution on the glacier surface, turbulent heat fluxes, and snow cover evolution on ice. The model is tested on glaciers situated in climatically and topographically contrasting regions—the Caucasus and Svalbard—using observational data for validation. The model is forced with ERA5 reanalysis data and employs morphometric glacial and topographic parameters. The simulated components of the surface energy and mass balance, as well as glacier dynamics over the period of 1984–2021, are presented. The model results demonstrate good agreement with observations, with correlation coefficients for accumulation, ablation, and total SMB ranging from 0.6 to 0.9. The primary driver of glacier retreat in the Caucasus is identified as an increase in net shortwave radiation balance caused by reduced cloudiness and albedo. In contrast, rapid glacier degradation in Svalbard is linked to an increased fraction of liquid precipitation and an extended snow-free period, leading to a sharp decrease in albedo. Full article

Senegal, like many West African countries reliant on natural resources and agriculture, faces severe impacts from climate change. This study provides an analysis undertaken by the United States Agency for International Development (USAID) under the Senegal Water Resources Management Activity, investigating historical and projected rainfall extremes to assess potential risks to water resources under future climate scenarios. Using bias-corrected CMIP6 data validated against the Enhancing National Climate Services (ENACTS) dataset for 1985–2014, we assess model performance through time series analysis, spatial distribution, and Taylor diagrams. We examine changes across three time periods—1985–2013 (historical), 2021–2040 (near future), and 2041–2060 (distant future)—focusing on nine key rainfall indices relevant to agriculture and water security. The results indicate that CMIP6 models capture historical rainfall patterns well. The models MPI-ESM1-2-HR, MIROC-ES2L, MRI-ESM2-0, CanESM5, and GISS-E2-1-G show the best performance and are recommended for climate impact assessments. Spatial analysis reveals prolonged dry periods in the north and heavier rainfall in the south. Under SSP585, the near future shows an increase in consecutive dry days (CDDs) and a decline in extreme rainfall events in northern Senegal, whereas the distant future projects a reversal with intensified rainfall (Rx5day). The south shows contrasting patterns, with increasing rainfall intensities in the long term. These findings highlight shifts in rainfall regimes and underscore the urgency of integrating future climate scenarios into adaptation planning. This study recommends extending analysis to temperature extremes due to their implications for agriculture and public health. Full article

This work reports on novel acid–base conjugate pairs of monocationic allyldiamidinium and dicationic diamidinium salts, some of which are ionic liquids (ILs) at ambient temperatures. A series of allyldiamidinium salts of the general formula [C₃H(NRMe)₄]X (R = Me, Et, Pr, allyl, CH₂CH₂OMe; X = Cl, bistriflimide, dicyanamide) were prepared from C₃Cl₄ or C₃Cl₅H and the appropriate secondary amine, RNMeH. Alkylated ethylenediamines similarly yield bicyclic allyldiamidinium salts, whereas longer diamines (H₂N(CH₂)_nNH₂ (n = 3, 4, 5)) were isolated as their conjugate acids, the diamidinium dicationic salts [C₃H₂(HN(CH₂)_nNH)₂]X₂. The salts were characterized by NMR, ES-MS, DSC, TGA, and miscibility or solubility studies. Additionally, the ILs were characterized by their viscosities. The conductivities of the diamidinium ILs were also measured, and this allowed for an investigation of their Walden parameters. In contrast to expectations, since the ion pairing and clustering were expected to be significant, this showed them to be “superionic”. Previous reports of Walden plots of dicationic ILs were found to be erroneous, and a reanalysis of the literature data found that all reported dicationic and even tetracationic ILs can be classified as superionic. The salts [C₃H(NMe₂)₄]Cl, [C₃H(EtN(CH₂)₂NEt)₂]OTf, and [C₃H₂(HN(CH₂)_nNH)₂]Cl₂ (n = 3, 4, 5) were also characterized by single-crystal X-ray diffraction. Full article

Accurate prediction of hemolytic peptides is essential for peptide safety evaluation and therapeutic design; however, existing models remain constrained by limited accuracy and interpretability. To overcome these challenges, we propose a regression framework that integrates embeddings from a protein language model with handcrafted amino acid descriptors. Specifically, sequence representations derived from the ESM2_t33 model are fused with physicochemical amino acid descriptor features, and key predictive variables are selected through a three-stage strategy involving variance filtering, F-test ranking, and mutual information analysis. The final ensemble model, composed of Random Forest, Extremely Randomized Trees, Gradient Boosting, eXtreme Gradient Boosting (XGBoost), and Ridge Regression, achieved a coefficient of determination (R²) of 0.57 and a correlation coefficient (R) of 0.76 on the test set, outperforming previous approaches. To enhance interpretability, we applied Shapley value analysis and the Calibrated_Explanation algorithm to quantify feature contributions and generate reliable sample-specific explanations. The trained model has been deployed online as HemPepPred, a tool for predicting hemolytic concentration (HC₅₀) values, which provides a practical platform for rational peptide design and safety assessment. Full article

Use of the upper limb to engage in everyday activities is a key indicator of functional recovery of stroke survivors. In addition to functional capacity, personal and environmental factors contribute to real-world upper limb use post-stroke. We aimed to combine data from the experience sampling method (ESM), a method used to capture real-time engagement in daily activities, with accelerometry, an objective measurement of arm use, to evaluate arm use behaviours of adult stroke survivors living in real-world environments. Thirty mild–moderately impaired stroke survivors and 30 age-standardized healthy individuals were monitored over 7 days, using accelerometers on both wrists and four ESM beeps per day to capture individual experiences in daily activities. Stroke survivors showed significantly lower use of the affected arm across all activity domains compared to the non-dominant arm of healthy participants and reported perceived lower skill and higher challenge levels. Physical context, motor capabilities and activity type were associated with affected arm use behaviour, with greater use observed during social settings and in physically demanding tasks. These findings demonstrate that combining ESM with accelerometry provides a novel, ecologically valid framework to capture and interpret the interplay between capacity, context, and behaviour in everyday life. This approach offers opportunities to design personalized, context-aware rehabilitation strategies that promote meaningful functional reintegration after stroke. Full article