The recurrent outbreak of viral pathogens and the possibility of the new pandemics demand the transition to the predictive and integrative computational frameworks instead of reactive one. This review describes computational antiviralism as an integrated approach that uses artificial intelligence (AI), virtual screening, and molecular design tools to identify antiviral targets at the molecular level. We compare deep learning-based structure models with physics-based molecular dynamics (MD) with network pharmacology to describe virus-host interactome dynamics. We also evaluate systems virology strategies that combine the transcriptomic, proteomic, and metabolomic data in order to solve infection-induced cellular reprogramming. The framework is not confined to the molecular, but includes evolutionary phylogenomics, epidemiological modelling of the zoonotic spillover, and climate-guided forecasting of cross-species transmission of viruses. We consider such key issues as assay heterogeneity, interpretability of models, and management of autonomous laboratory systems. Importantly, we explicitly acknowledge that no true end-to-end validated multi-scale antiviral pipeline currently exists; the framework is presented as a forward-looking research agenda with clearly defined open challenges. Collectively, this synthesis will bring computational antiviralism as an anticipatory field of study that can catalyze the broad-spectrum antiviral discovery, as well as providing preemptive countermeasures to emergent viral challenges through coordinated molecular, cellular and ecosystem-level interventions.
Single-cell RNA sequencing (scRNA-seq) has transformed our understanding of tumor heterogeneity, yet the impact of computational pipeline choices on biological conclusions remains poorly characterized. Here, we systematically benchmark 465 combinations of 5 normalization methods and 4 clustering algorithms across 3 cancer datasets encompassing over 434,000 cells. We introduce the Biological Discordance Score (BDS), a metric that quantifies how marker gene interpretation changes across pipelines. We find that normalization method choice has a greater impact on biological interpretation than clustering algorithm selection. Strikingly, pipelines with similar clustering agreement (ARI) can identify up to 86% different marker genes, a discrepancy invisible to standard evaluation metrics. Log-normalization consistently achieves the best balance of performance and interpretive stability across cancer types. Using a Borda count meta-ranking framework with bootstrap confidence intervals, we provide unified recommendations for pipeline selection. Our results demonstrate that computational choices have profound but underappreciated consequences for biological discovery in cancer single-cell studies.
Spore-forming bacteria such as Bacillus cereus pose a significant public health challenge due to their ability to survive harsh environmental conditions, resist conventional decontamination strategies, and cause recurrent infections in clinical and food-associated environments. This persistence is primarily associated with sporulation, a tightly regulated developmental process in which long-term survival depends on maintaining genome integrity while cellular metabolism remains largely inactive. DNA-binding proteins are therefore central to sporulation, as effective DNA condensation and protection are essential for sporulation progression and cyst wall maturation. Among these, α/β-class small acid-soluble proteins (SASPs), particularly SASP2, bind to the DNA minor groove and stabilize the genome during dormancy. In this study, a structural model of the B. cereus SASP2-DNA complex was constructed and analyzed through an integrated computational approach to identify compounds targeting this interaction. Phytochemicals derived from Garcinia mangostana, Garcinia cowa, Ficus exasperata, and Entada abyssinica, previously reported for antimicrobial activity, were evaluated for their potential to interact with the SASP2-DNA interface, a mechanism not previously explored. Several compounds showed strong binding affinity at the SASP2-DNA minor-groove interface and were predicted to influence key interactions under simulated stress conditions, leading to DNA compaction stability and stress tolerance, which may subsequently affect cyst wall formation and spore viability. Notably, the identification of plant-derived compounds capable of targeting the SASP2-DNA interface represents a novel observation. Overall, these findings provide a promising computational basis for exploring strategies to limit the persistence and transmission of B. cereus infections.
This study investigated the effects of glycyrrhizic acid and liquiritin-the primary bioactive compounds in Glycyrrhiza uralensis Fisch.-on myosin conformational dynamics, antioxidative functionality, and polycyclic aromatic hydrocarbons (PAHs) generation. Both phytochemicals interacted with myosin via static quenching mechanisms, eliciting structural rearrangements that amplified radical scavenging efficacy, retarded oxidation, and ultimately suppressed PAHs accumulation. Quantum chemical calculation revealed glycyrrhizic acid's diminished HOMO-LUMO energy gap compared to liquiritin, signifying superior electron-donating capacity and molecular affinity. Subsequent molecular dynamics simulations corroborated heightened binding stability for glycyrrhizic acid, characterized by persistent interactions at residues Asp472, Phe471, and Glu469. Collectively, these findings elucidate a structural modulation mechanism wherein glycyrrhizic acid enhances myosin's antioxidative defense to attenuate PAHs formation, advancing natural intervention paradigms for optimizing safety and nutritional quality in protein-based food matrices.
Environmental exposure to the monochloroacetic acid (MCA), a disinfection by-product, poses a serious public health crisis, particularly through its interaction with skin barrier-associated proteins. In this study, a spectroscopy-based integrative strategy was employed to investigate the interaction between MCA and transglutaminase (TGase), a key enzyme involved in epidermal barrier integrity. UV-vis absorption, steady-state, time-resolved, three-dimensional, and synchronous fluorescence spectroscopy analyses, revealed that MCA bound to TGase predominantly via hydrogen bonding and van der Waals forces, following a static quenching mechanism. In addition, MCA altered the local microenvironment of TGase and induced conformational change. Thermodynamic analysis showed negative values of the free energy at different temperatures, indicating that the binding process was spontaneous. Molecular docking suggested that MCA binding induced conformational changes in TGase, likely due to interactions with the catalytic residues Asp255 and His274 within the active site. Furthermore, molecular dynamics simulations revealed that MCA binding enhanced the local conformational flexibility of TGase. Complementary in vitro assays linked these structural perturbations to altered TGase activity and impaired epidermal barrier function. This study elucidates the structure-function relationship underlying the MCA-TGase interaction and provides molecular-level insights into the potential dermal toxicity of MCA.
Interoceptive interventions offer a promising avenue for improving mental health conditions, which commonly feature bodily or interoceptive symptoms. Perceptual accuracy for interoceptive signals, such as heartbeats, varies across individuals and presents a potential target for treatment. Adult participants (N = 28, 20F) completed eight sessions of a cardiac interoception training protocol, and their anxiety reduction was compared to that in a passive control group (N = 26, 22F). Bayesian computational models were compared to identify mechanisms of perception and learning that best explained participants' responses during the heartbeat discrimination task. Parameter estimates from the best-fitting model were used as computational phenotypes to explain anxiety reduction due to interoceptive training. Interoceptive training improved perceptual accuracy in two tasks of heartbeat perception and reduced self-reported trait anxiety. Computational modelling indicated that accuracy improvement in the heartbeat discrimination task was explained by increases in the internal reliability estimate for interoceptive signals - their precision weighting - while a lower-level parameter representing stable sensory noise moderated this precision weighting improvement by influencing the speed of learning. Reductions in both state and trait anxiety scores in the training group were uniquely explained by computational parameter estimates, and not by conventional accuracy measures. These findings indicate that cardiac interoceptive accuracy is modifiable and can be targeted to alleviate anxiety symptoms, and that interoceptive interventions may be best guided by a computational phenotyping approach.
Early and accurate diagnosis of Alzheimer's Disease (AD) is crucial for effective clinical intervention. In this study, we propose a lightweight vision transformer architecture specifically designed for AD classification using 2D brain MRI slices. LICAUN-ViT incorporates three key innovations: Mono-Head Self-Attention (MOHSA) to reduce computational overhead, Uniformity Normalization (Uni-Norm) to mitigate oversmoothing and enhance feature diversity, and Context-Aware Convolution (CAC) to integrate long-range dependencies with local structural features. Evaluated on two benchmark datasets derived from the Alzheimer's Disease Neuroimaging Initiative (ADNI), our model achieves state-of-the-art performance with an accuracy of 93.03 % on axial slices and 94.15 % on sagittal slices, while maintaining relatively low floating-point operations (FLOPs) for efficient deployment. Extensive ablation studies and singular value analyses confirm the effectiveness and robustness of the proposed components. These results demonstrate that the proposed model offers a computationally efficient and promising solution for automated AD diagnosis, with strong potential for clinical integration.
Few studies have reported that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein may suppress cancer cell growth. Here, we investigated the effect of the SARS-CoV-2 spike protein in A549 lung cancer cells using recombinant spike protein (SP), spike protein transfection (SPT), and pseudo-SARS-CoV-2 virus (PSV). To evaluate its anticancer effects and associated molecular changes, we performed RNA sequencing, colony formation, immunofluorescence, cell viability, migration, FACS, western blotting, 3D spheroid, molecular docking, siRNA knockdown, qRT-PCR, apoptosis assays, and computational interaction analyses. Spike protein significantly inhibited the long-term growth of A549 cells. Among the delivery methods, PSV showed the most potent anticancer effect, followed by SPT and SP, as evidenced by reduced migration, spheroid growth, and increased sub-G1 arrest. RNA-Seq identified two differentially expressed lncRNAs, with MEG3 upregulated and BCYRN1 downregulated following spike-related treatment. Functional experiments showed that BCYRN1 knockdown or MEG3 overexpression reproduced key spike protein-induced antiproliferative and pro-apoptotic effects, including increased cleaved caspase-3 expression. Computational analyses suggested possible interactions between the Spike protein and these lncRNAs. In addition, actinomycin D (ActD) chase experiments indicated altered transcript dynamics of MEG3 and BCYRN1 under spike-related conditions. Collectively, these findings suggest that SARS-CoV-2 spike protein exerts antitumor activity in A549 cells and may contribute to the regulation of MEG3 and BCYRN1. Further studies, including formal rescue and biochemical interaction assay, will be required to establish causality and direct molecular mechanisms.
Exposure assessment based on proximity buffers is widely used in environmental epidemiology. However, the conventional workflow of polygon buffering and spatial intersection is computationally intensive for large datasets and repeated assessments over time. We introduce grid-based approximation and vectorized extraction (GAVE) to reduce the computational burden of buffer-based exposure assessment. We describe the GAVE workflow and benchmark its performance by estimating the average dust-specific fine particulate matter (PM2.5) within 2-15-km circular buffers around 1,000 random locations in California during biweekly exposure periods, comparing it with the conventional workflow and raster-based methods. We demonstrate the scalability of GAVE by applying it to estimate prenatal dust exposures for 7 million individuals in the Study of Outcomes in Mothers and Infants longitudinal administrative birth cohort. GAVE achieved a 366-fold speed improvement with negligible difference in exposure estimates (correlation > 0.999) compared to the conventional workflow. It supports simultaneous calculations of multiple buffer sizes and exposure time frames, avoids rasterization, and can be implemented entirely in R. This open-source implementation technique for buffer-based exposure assessment provides a practical alternative to conventional buffer analysis and raster-based extraction tools, especially in large datasets with longitudinal exposure timeframes.
Interactomics aims at identifying and characterizing in a comprehensive way the complete set of protein-protein interactions that occur within a cell, tissue, or organism. In this chapter, we cover the most relevant experimental and computational approaches that are used in the field to assess the interactome of a specific protein or of all the proteins in a defined compartment. We address their applicability as well as their advantages and disadvantages. We then describe the different databases available to researchers interested in dissecting protein-protein interactions and finally briefly summarize the several attempts that have been made to identify the human interactome, i.e., the complete set of protein interactions that occur within a human cell.
Conventional treatment methods struggle to effectively eliminate sulfamethoxazole (SMX) from wastewater due to its persistent aromatic structure and sulfonamide moiety, while existing advanced oxidation processes often suffer from high energy consumption and inadequate degradation pathway control. This study demonstrates an advanced approach for SMX degradation in aqueous systems through dielectric barrier discharge (DBD) plasma technology synergistically integrated with COMSOL multiphysics simulation and particle swarm optimization-support vector machine (PSO-SVM) modeling. Computational simulations revealed that the DBD plasma generates a diffuse arc (D-A) mode electric field with Penning ionization effects, where primary reactive species formation predominantly occurs within the sheath layer. Under optimized operational parameters (input power: 50 W, pH 7, conductivity: 3.16 μS/cm, initial concentration: 10 mg/L, the carrier gas: air), the system achieved an exceptional SMX removal efficiency of 99.82 %, corresponding to a constant reaction rate of 0.3087 min-1. The machine learning-optimized process simultaneously enhanced energy yield by 18.11 mg/(kW·h) and improved energy utilization efficiency by 15.11 %. Notably, the system maintained robust performance (>75 % SMX removal) in the presence of competing inorganic anions (NO3-, HCO3- and Cl-). Mechanistic investigations combining DFT, EPR and UPLC-MS/MS elucidated that •OH, O2•-, 1O2, •NO2 and ONOOH preferentially attack the electrophilic N(4), N(7), and N(17) sites, initiating sequential diazotization and nitration pathways. Practical validation using hospital wastewater demonstrated significant removal of COD (56.17 %) and TOC (40.16 %), while effectively suppressing antimicrobial resistance genes. This work establishes PSO-SVM optimized DBD plasma as a sustainable and intelligent solution for pharmaceutical wastewater remediation.
Metabolomics has emerged as a powerful discipline for capturing dynamic biochemical states and linking molecular processes to human health, disease, and biotechnology. This chapter explores the evolution of metabolomics from early profiling efforts to its current integration with advanced analytical platforms, computational biology, and precision medicine. We highlight how targeted and untargeted metabolomic strategies have advanced biomarker discovery, clarified disease mechanisms, and informed the development of personalized therapies in cancer, cardiovascular disease, diabetes, and other conditions. The role of pharmacometabolomics in refining drug response predictions and guiding therapeutic decisions is also examined. Beyond biomedicine, metabolomics drives innovation in biotechnology, synthetic biology, and bio-based production, while emerging tools such as single-cell metabolomics and machine learning continue to expand its scope. By directly tracing metabolic fluxes and engineering new pathways, metabolomics is becoming a practical engine for designing microbes, optimizing bioprocesses, and accelerating drug discovery.
To address the limitations of traditional multi-rigid-body models in accurately analyzing the vibration characteristics of rack-climbing lift platforms, this study proposes a rigid-flexible coupled dynamic equation and develops a flexible-body substructure reduction method to enhance simulation accuracy while reducing computational complexity. A rigid-flexible coupled simulation model is constructed by integrating ANSYS APDL with ADAMS, enabling an organic coupling between flexible components and rigid bodies. Axial vibration characteristics of the platform are simulated using four models-rigid, multi-flexible coupled, flexible-driven coupled, and rigid-flexible coupled. Results indicate that the Z-direction vibration is similar to the X-direction, with maximum amplitudes of 1.03mm and 1.34mm occurring during the startup phase, whereas the Y-direction amplitude reaches 7mm, representing increases of 579% and 422% relative to the Z- and X-directions, respectively. These findings demonstrate the inadequacy of rigid-body models for effective vibration analysis. Compared with the multi-flexible and flexible-driven coupled models, the rigid-flexible coupled model exhibits smoother transition in vibration response and more pronounced amplitudes, confirming its validity. Further parametric analysis under varying operating speed, load magnitude, eccentric loading, and braking time reveals that eccentric loading effects are condition-dependent, while load magnitude and operating speed significantly influence platform vibration. Finally, experimental validation of axial vibration and vibration acceleration confirms the accuracy of the proposed model. This study provides a theoretical foundation and reliable modeling framework for structural optimization, comfort design, and vibration behavior analysis of lift platforms.
Protein secondary structure prediction represents an important intermediate step between a protein's linear amino acid sequence and its three-dimensional structure, with broad implications for synthetic biology, drug development, and disease research. Although experimental techniques such as X-ray crystallography provide highly accurate structural information, they are labor-intensive, time-consuming, and costly, which has motivated the development of computational alternatives. Early machine-learning approaches to this problem were limited in their ability to capture complex sequence-structure relationships. The introduction of convolutional and recurrent neural networks improved hierarchical feature extraction, and predictive performance advanced further with transformer-based architectures such as AlphaFold2. This review outlines recent advances in hybrid model design, benchmark datasets, and evaluation metrics for protein secondary structure prediction. We also discuss current methodological limitations, including data dependency and dataset bias, and outline future directions such as cross-species validation, uncertainty-aware modeling, and the still-emerging potential of incorporating heterogeneous biological data into next-generation PSSP frameworks.
Mass spectrometry (MS)-based metabolomics is a powerful tool for understanding the complexity of biochemical processes and to identify biomarkers across diverse biological systems. The vast amount of data generated by extreme resolution mass spectrometers poses significant data processing challenges, requiring robust computational approaches and workflows for meaningful data interpretation. This chapter provides a comprehensive overview of current methodologies in MS-based metabolomics data analysis, with a focus on data preprocessing and pretreatment, m/z extraction and annotation, univariate and multivariate statistical approaches, as well as data visualization. We discuss key considerations for ensuring data quality and the growing role of bioinformatics in pathway analysis and metabolite identification. We highlight the transforming role of extreme resolution and mass accuracy enabled by FT-ICR mass spectrometers, and finally, we explore emerging trends, including artificial intelligence-driven insights and real-time data processing, to guide future developments in this rapidly evolving field.
Analysis of omics experiments usually identifies lists of molecular entities (genes, transcripts, other RNA molecules, proteins, or metabolites) that need to be functionally interpreted, not as individual molecules but as ensembles that are functionally related. Multiple computational methods have been developed to study the functions or pathways that are more relevant in those lists of biomolecules. In this chapter, we describe the principles behind these methods and discuss their use in the functional interpretation of omics data. First, we describe the two main approaches to identify the most relevant functional annotations in a list of biomolecules: Over-Representation Analysis (ORA) and Functional Class Scoring (FCS). Afterward, we introduce basic concepts of network biology and discuss how network analysis methods, such as Centrality Analysis, Network Clustering, and Network diffusion, can be used to potentiate the functional interpretation of omics data.
In this work, a series of deep eutectic solvents (DESs) were designed, using diamines as hydrogen bond acceptors (HBAs) and phenol derivatives as hydrogen bond donors (HBDs). The DES composed of 1,3-propanediamine and m-cresol was screened as the most efficient system for astaxanthin recovery from Haematococcus pluvialis. This dual-functional DES enables the one-step extraction and conversion of astaxanthin esters to astaxanthin. The structure of the DES was characterized by ¹H NMR, while computational chemistry confirmed the OH⋯N hydrogen bond within the DES. Quantum chemical calculations further clarified the conversion mechanism, demonstrating DES-catalyzed ester bond cleavage in astaxanthin esters to form free astaxanthin. Furthermore, optimizations of cell disruption and extraction processes improved the recovery efficiency, achieving a yield of 35.8 ± 0.13mg/g under optimal conditions (molar ratio 3:1, extraction time 2.1h, and material-to-liquid ratio 1:110g/mL). Overall, this strategy offers an efficient, eco-friendly approach for the direct production of free astaxanthin from Haematococcus pluvialis.
Aptamers, as single-stranded DNA (ssDNA) or RNA oligonucleotides, are pivotal in biosensing due to their high affinity. However, excessive lengths of these nucleic acid probes can impair their binding affinity and target recognition efficiency. Traditional optimization methods, such as static structural modeling, fail to capture the dynamic interactions between aptamers and biological macromolecules. Therefore, optimizing aptamer length to enhance affinity while maintaining effective target recognition is crucial. Here, we employed 600ns molecular dynamics (MD) simulations using the amberff14sb and parmbsc1 force fields, alongside molecular mechanics/generalized Born surface area (MM/GBSA) free energy calculations to optimize the binding affinity of a ssDNA aptamer targeting the hemagglutinin-neuraminidase (HN) protein-a critical surface receptor of Newcastle disease virus (NDV) responsible for viral attachment and entry. By systematically truncating the aptamer sequence guided by normalized criteria to eliminate length bias, we identified a 10-nucleotide variant (fqh-2) that exhibited a hydrogen bond efficiency ratio (HBER) of 1.055 and a binding free energy efficiency ratio (BFEER) of -5.124 kcal/mol, reflecting enhanced interactions with the HN protein. Furthermore, a graphene oxide (GO)-based fluorescence quenching assay confirmed a threefold increase in binding affinity for the optimized aptamer, aligning with computational predictions. This study not only provides a dynamic structure-guided framework for aptamer optimization but also lays a theoretical foundation for further advancements in optimizing and tailoring aptamers for specific applications.
PROTACs, also called proximity-inducing agents, are chimeric molecules composed of a ligand for protein of interest (POI), an E3 ligase ligand and a linker connecting them. PROTACs have transformed the therapeutic landscape by enabling an event-driven strategy to degrade disease-associated proteins previously regarded as undruggable. The unique event-driven mechanism of PROTACs allows selective protein degradation with greater potency and lower drug resistance than conventional occupancy-based inhibitors. Despite their advantages, challenges such as high molecular weight, low permeability, poor pharmacokinetic properties restrict their clinical applications. To overcome these limitations, AI-driven technologies are being utilised to generate novel, chemically valid PROTACs. This review highlights the drawbacks of conventional computational methods and explores emerging AI-driven tools applied to multiple areas of PROTAC research, such as target (POI) selection (DeepUSI, DrugnomeAI), linker generation (AIMLinker, DiffLinker), activity prediction (AI-DPAPT, DeepPROTAC), POI degradability assessment (PrePROTAC, MAPD), ternary complex modelling (ProFlow), PROTAC generation (PROTAC-RL), and ADME property estimation (MT-GNN). It also outlines current challenges such as data scarcity, reproducibility issues, inadequate model generalizability, emphasizing the need for hybrid models or integrated AI techniques to mitigate these limitations.
This study aims to systematically map and analyze the knowledge structure, research hotspots, and evolutionary trends in the field of Type B Aortic Dissection (TBAD) using bibliometric methods. CiteSpace and VOSviewer were employed to conduct quantitative and visual analyses across multiple dimensions, including publication trends, collaboration networks among countries/institutions/authors, keyword co-occurrence and evolution, as well as co-citation analysis and reference burst detection. The number of publications in this field has shown an overall upward trend, with a significant increase since 2013, reflecting the growing clinical and scientific attention to this life-threatening vascular emergency. The United States (USA) and China form a dual-core global collaboration structure, with the USA leading in international cooperation depth and citation impact, and China emerging as a major contributor with rapid growth in research output - an evolution that mirrors the global redistribution of cardiovascular research capacity. Keyword analysis reveals a paradigm shift from technical application to precision intervention and individualized management, while current research focusing on Thoracic Endovascular Aortic Repair (TEVAR), complication prediction, computational fluid dynamics (CFD), and artificial intelligence (AI)-assisted diagnosis and treatment. Co-citation analysis confirms TEVAR as the gold-standard minimally invasive treatment for TBAD, with its widespread acceptance driving the standardization of clinical practice. Burst analysis of keywords indicates that "prediction model" and "deep learning" have become emerging research hotspots, marking the entry of TBAD research into an intelligent, data-driven era. Research on TBAD has developed a sophisticated knowledge system over the past two decades, shifting from traditional surgical exploration to an intelligent, data-driven research paradigm. This bibliometric analysis identifies a USA-China dual-core global collaboration pattern in the field and a three-stage evolution of research focus from pathophysiological exploration to evidence-based TEVAR standardization, and further to the integration of CFD and AI. Critical research gaps are also highlighted, including under-investigation of high-risk populations, insufficient long-term evidence for TEVAR, and inadequate cross-disciplinary integration of CFD, AI and genomics. Future TBAD research should prioritize multicenter prospective trials to upgrade clinical evidence, advance interdisciplinary precision medicine models, and build globally standardized big data platforms for the development and validation of AI-based diagnostic and therapeutic tools, thereby achieving more scientific and personalized management of TBAD.