Search Results (2,942)

Depression remains a leading cause of global disability, yet scalable and objective diagnostic tools are still lacking. Speech has emerged as a promising non-invasive modality for automated depression detection, due to its strong correlation with emotional state and ease of acquisition. While convolutional neural networks (CNNs) have achieved state-of-the-art performance in this domain, their high computational demands limit deployment in low-resource or real-time settings. Spiking neural networks (SNNs), by contrast, offer energy-efficient, event-driven computation inspired by biological neurons, but they are difficult to train directly and often exhibit degraded performance on complex tasks. This study investigates whether CNNs trained on audio data from the clinically annotated DAIC-WOZ dataset can be effectively converted into SNNs while preserving diagnostic accuracy. We evaluate multiple conversion thresholds using the SpikingJelly framework and find that the 99.9% mode yields an SNN that matches the original CNN in both accuracy (82.5%) and macro F1 score (0.8254). Lower threshold settings offer increased sensitivity to depressive speech at the cost of overall accuracy, while naïve conversion strategies result in significant performance loss. These findings support the feasibility of CNN-to-SNN conversion for real-world mental health applications and underscore the importance of precise calibration in achieving clinically meaningful results. Full article

Background: Anaplastic lymphoma kinase (ALK) is a key receptor tyrosine kinase involved in regulating signaling pathways critical for cell proliferation, differentiation, and survival. Mutations or rearrangements of the ALK gene lead to aberrant kinase activation, driving tumorigenesis in various cancers. Although ALK inhibitors have shown clinical benefits, drug resistance remains a significant barrier to long-term efficacy. Developing novel ALK inhibitors capable of overcoming resistance is therefore essential. Methods: A structure-based pharmacophore model was constructed using the 3D structures of five approved ALK inhibitors. Systematic virtual screening of the Topscience drug-like database was performed incorporating PAINS filtering, ADMET prediction, and molecular docking to identify promising candidates. In vitro antiproliferative assays, molecular docking, molecular dynamics simulations, and MM/GBSA binding free energy calculations were used to evaluate biological activity and elucidate binding mechanisms. Results: Two candidates, F1739-0081 and F2571-0016, were identified. F1739-0081 exhibited moderate antiproliferative activity against the A549 cell line, suggesting potential for further optimization. Computational analyses revealed its probable binding modes and interactions with ALK, supporting the observed activity. Conclusions: This study successfully identified novel ALK inhibitor candidates with promising biological activity. The integrated computational and experimental approach provides valuable insights for the rational design of optimized ALK inhibitors to address drug resistance in cancer therapy. Full article

The latent class model has been proposed as a powerful tool in understanding human behavior for various fields such as social, psychological, behavioral, and biological sciences. However, one important limitation of the latent class model is that it is primarily applied to data with binary responses or categorical responses, making it fail to model real-world data with continuous or negative responses. In many applications, ignoring the weights throws out a lot of potentially valuable information contained in the weights. To address this limitation, we propose a novel generative model, the arbitrary-distribution latent class model (adLCM). Our model enables the generation of data’s response matrix from an arbitrary distribution with a latent class structure. When compared to the latent class model, our adLCM is both more realistic and general. To our knowledge, our adLCM is the first model for latent class analysis with any real-valued responses, including continuous, negative, and signed values, thereby extending the classical latent class model beyond its traditional limitation to binary or categorical outcomes. We investigate the identifiability of the model and propose an efficient algorithm for estimating the latent classes and other model parameters. We show that the proposed algorithm enjoys consistent estimation. The performance of our algorithm is evaluated using both computer-generated data and real-world personality test data. Full article

Endophytes are important sources of bioactive secondary metabolites with therapeutic and agricultural relevance. This study reports the isolation and characterization of bioactive compounds from endophytic Fusarium solani associated with Solanum surattense. The fungal strain, selected after preliminary screening for its antimicrobial potential, was identified through morphological and molecular methods. A pure compound, 4-hydroxyphenyl 8-chlorooctanoate with a molecular mass of 270, was obtained and structurally characterized using GC–MS, FTIR, and NMR spectroscopy. Its anti-microbial potential was evaluated through molecular docking against key bacterial (Staphylococcus aureus) and fungal (Aspergillus fumigatus) targets, showing notable binding affinities with ClpP protease (−7.1 kcal/mol) and 14α-demethylase (−7.4 kcal/mol), respectively. Molecular dynamics simulations further confirmed the stability of the 5FRB-compound complex, with lower RMSD and RMSF values indicating strong structural integrity. Supporting analyses (B-factor and radius of gyration) confirmed the compactness and rigidity of the complex. These findings highlight the potential of 4-hydroxyphenyl 8-chlorooctanoate as a promising antimicrobial agent and provide a strong basis for further in vitro and in vivo validation of the purified compound as an antimicrobial candidate. Full article

The high-dimensional chaos generated in a neural network consisting of pseudo-neuron devices invented by one of the authors (S.N.) has been successfully applied to control the complex motion of a roving robot, e.g., to solve a maze, as reported in the previous papers. On the basis of successful works and the concept that chaos plays important functional roles in biological systems, in the present paper, we report new experiments to show the functional aspects of chaos via behavioral interactions in an ill-posed context and solve problems with chaotic neural networks. Explicitly, experiments on two roving robots in a maze (labyrinth) are reported, in which both seek to catch each other or one chases and the other flees, mimicking the survival activities of insects in natural environments. The two-dimensional robot motion is controlled with motion control systems, each of which is equipped with a chaotic neural network to generate autonomous and adaptive actions depending on sensor inputs of obstacles and/or target detection information including uncertainty. We report both computer experiments and practical hardware implementations, where for the latter, only the chaotic neural network is run on a desktop computer, the motion signals are coded into two-dimensional space, and sensor signals are transferred via Bluetooth device between robots and computers. Full article

According to a recent experimental phenomenology–information processing theory, the sensory strength, or vividness, of visual mental images self-reported by human observers reflects the intensive variation in subjective time duration during the process of generation of said mental imagery. The primary objective of this study was to test the hypothesis that a biologically plausible essential multilayer perceptron (MLP) architecture can validly map the phenomenological categories of subjective time duration onto levels of subjectively self-reported vividness. A secondary objective was to explore whether this type of neural network cognitive modeling approach can give insight into plausible underlying large-scale brain dynamics. To achieve these objectives, vividness self-reports and reaction times from a previously collected database were reanalyzed using multilayered perceptron network models. The input layer consisted of six levels representing vividness self-reports and a reaction time cofactor. A single hidden layer consisted of three nodes representing the salience, task positive, and default mode networks. The output layer consisted of five levels representing Vittorio Benussi’s subjective time categories. Across different models of networks, Benussi’s subjective time categories (Level 1 = very brief, 2 = brief, 3 = present, 4 = long, 5 = very long) were predicted by visual imagery vividness level 1 (=no image) to 5 (=very vivid) with over 90% success in classification accuracy, precision, recall, and F1-score. This accuracy level was maintained after 5-fold cross validation. Linear regressions, Welch’s t-test for independent coefficients, and Pearson’s correlation analysis were applied to the resulting hidden node weight vectors, obtaining evidence for strong correlation and anticorrelation between nodes. This study successfully mapped Benussi’s five levels of subjective time categories onto the activation patterns of a simple MLP, providing a novel computational framework for experimental phenomenology. Our results revealed structured, complex dynamics between the task positive network (TPN), the default mode network (DMN), and the salience network (SN), suggesting that the neural mechanisms underlying temporal consciousness involve flexible network interactions beyond the traditional triple network model. Full article

Immunotherapy with immune checkpoint inhibitors has changed the treatment landscape in many solid tumors. Despite the unprecedent success, many patients will develop primary or secondary resistance to treatment or will hold up therapy due to the emerging immune-related toxicity. Traditionally, tissue-based immune biomarkers, such as PD-L1 expression, have been used to select patients who will benefit most from immunotherapy. However, these markers demonstrate major limitations, such as tumor heterogeneity and sample constraints. In addition, they do not reflect the dynamic interplay of tumor and hosts immune response during treatment. Peripheral blood immunomarkers offer a minimally invasive, real-time assessment of the immune system and its interaction with the tumor. Integration of traditional tissue-based and peripheral blood markers coupled with the recent developments in computational platforms, artificial intelligence, and machine learning models may provide more successful biomarkers for prognosis, prediction of immunotherapy-related outcomes, the early evaluation of forthcoming disease progression, and the prediction of the emerging immune-related adverse events. Despite the promising developments in the field of immune biomarkers, several issues including assay standardization, clinical validation, and biological variability should be addressed to improve personalized immunotherapy approaches. In this comprehensive review we provide an update on immune biomarker evolution, and we discuss the current limitations and future directions. Full article

Background/Objectives: The current standard treatment for locally advanced rectal cancer (LARC) is neoadjuvant chemoradiotherapy, or total neoadjuvant therapy (TNT), followed by total mesorectal excision (TME). If the neoadjuvant treatment results in a clinical complete response (cCR), non-operative management of LARC might be possible. It is hypothesized that cCR rates will increase with increasing radiotherapy doses. By using proton therapy, doses to organs at risk (OAR) may be decreased. In preparation for a clinical trial on dose-escalated proton therapy for LARC, the purpose of this study is to establish the feasibility of proton therapy for dose-escalated hypofractionated radiotherapy of LARC. Methods: Ten patients, having previously received short course radiotherapy (SCRT) for LARC, were included in this planning study. Two photon plans and two proton plans were created for each patient: one with a standard 5 × 5 Gy fractionation and one dose-escalated up to 5 × 7 Gy. Proton plans were robustly optimized. For all plans the integral dose (ID) was computed, and for the proton plans relative biological effectiveness (RBE) distributions were calculated. Feasibility was assessed in terms of target coverage and OAR doses. Results: All treatment plans satisfied target coverage criteria. Three of the photon and two of the proton dose-escalated plans exceeded recommended OAR objectives. Proton IDs were on average lower by a factor of 1.97 compared to photon IDs. Mean doses to OAR were, in general, lower for protons. All proton RBE values in the escalated target volumes were between 1.09 and 1.16. Conclusions: The proposed dose escalation was found to be feasible. Protons can reduce the integral dose and mean doses to OARs compared to photons in both the dose-escalated and non-escalated cases. Differences in RBE between escalated and standard fractionation were small. Full article

Phytoecdysteroids represent a class of naturally occurring substances known for their diverse biological functions, particularly their strong ability to stimulate protein anabolism. In this study, a computational machine learning-driven quantitative structure–activity relationship (QSAR) approach was applied to analyze the anabolic potential of 23 ecdysteroid compounds. The ML-based QSAR modeling was conducted using a combined approach that integrates Genetic Algorithm-based feature selection with Multiple Linear Regression Analysis (GA-MLRA). Additionally, structure optimization by semi-empirical quantum-chemical method was employed to determine the most stable molecular conformations and to calculate an additional set of structural and electronic descriptors. The most effective QSAR models for describing the anabolic activity of the investigated ecdysteroids were developed and validated. The proposed best model demonstrates both strong statistical relevance and high predictive performance. The predictive performance of the resulting models was confirmed by an external test set based on R²_test values, which were within the range of 0.89 to 0.97. Full article

Asthma has traditionally been viewed as a single disease, but recent research reveals its clinical and molecular complexity. This perspective highlights the need to shift from a traditional, uniform treatment paradigm to one that embraces the heterogeneity of asthma across individuals. Each patient presents a unique clinical story shaped by a complex interplay of genetic predispositions, developmental programming during critical early-life windows, the influence of sex and hormones, and lifelong environmental exposures. Asthma comprises multiple subtypes with distinct clinical and biological features. Furthermore, lifestyle factors such as obesity and smoking, along with highly prevalent comorbidities like allergic rhinitis and gastroesophageal reflux disease, significantly modify the disease’s course and response to treatment. This article explores how classifying the disease into clinical phenotypes (observable characteristics) and molecular endotypes (underlying mechanisms)—particularly the distinction between T2-high and T2-low inflammation—provides a crucial framework for managing this complexity. The application of this framework, guided by biomarkers, has enabled the development of targeted biologic therapies that can transform care for specific patient subgroups. Despite these advances, significant challenges remain. The pathophysiology of certain subgroups, particularly non-T2 asthma, remains poorly defined, and there is an urgent need for reliable predictive biomarkers to guide therapy and monitor outcomes. It is our opinion that future studies must adopt a systems-biology strategy, with a multi-omics approach that constructs a comprehensive molecular profile of each patient. This integrative methodology will require the use of advanced computational methods, including machine learning and artificial intelligence, to decipher the complex pathways linking genetic and environmental inputs to clinical disease. In conclusion, this article argues for a more personalized understanding of asthma, urging clinicians and researchers to consider each patient’s unique clinical presentation. Full article

Mitochondrial dysfunction is implicated across a spectrum of neurological diseases, yet its causal role and mechanistic specificity remain unclear. This study employed a multi-modal integrative analysis of mitochondrial gene expression in Alzheimer’s Disease (AD), Amyotrophic Lateral Sclerosis (ALS), Multiple Sclerosis (MS), and Parkinson’s Disease (PD) to address these gaps. We combined machine learning for predictive modeling with genetic causal inference methods (Mendelian Randomization, colocalization, PheWAS), followed by drug enrichment analysis and molecular docking. Our machine learning models, particularly Support Vector Machine and Multi-layer Perceptron, effectively classified these conditions, with MS exhibiting the highest predictability (mean Accuracy: 0.758). Causal inference analyses identified specific gene–disease links; for instance, genetically predicted increased expression of PDK1 was causally associated with an elevated risk for both AD (OR = 1.041) and ALS (OR = 1.037), identifying pyruvate metabolism as a shared vulnerability. In contrast, genes like SLC25A38 emerged as highly predictive specifically for PD. We also observed evidence of potential brain–periphery interaction, such as a bidirectional causal relationship between red blood cell indices and MS risk. Finally, drug enrichment analysis highlighted Celecoxib, and subsequent molecular docking predicted a strong binding affinity to PDK1 (docking score S = −6.522 kcal/mol), generating hypotheses for potential metabolic modulation. Taken together, this study provides a computational hypothesis framework suggesting mitochondrial pathways and targets that warrant future biological validation. This study provides specific, genetically supported evidence for the causal role of mitochondrial pathways in neurological diseases and identifies tangible targets for future therapeutic development. Full article

Cancer is a multifaceted and life-threatening disease characterized by the unregulated proliferation of malignant cells. Developing new therapies and diagnostic methods for cancer remains a critical focus of research. Proteins involved in cancer progression are being targeted to facilitate the discovery of effective biological treatments. Among these, the ART1 protein plays a critical role in promoting cancer progression, establishing it as a key target for drug therapy. Actinomycetes, known for their anticancer activity, were explored in this study for their potential to inhibit ART1. One hundred bioactive secondary metabolites derived from actinomycetes were subjected to in silico screening to evaluate their potential anticancer activity through inhibition of ART1. The three-dimensional structure of ART1 was generated using the SWISS-MODEL tool and validated through the Save server 6.0 and ProSa web. The structural stability of the ART1 protein was evaluated through molecular dynamics analysis using the iMod server. The potential active sites within the ART1 structure were mapped using the Computed Atlas of Surface Topography of Proteins (CASTp). Molecular docking and protein–ligand interaction studies were performed using AutoDock Vina. Additionally, pharmacophore modeling was conducted using the Pharmit server to identify promising compounds. Toxicity predictions and in silico drug-likeness assessments were carried out using Swiss-ADME and ADMET Lab which evaluate Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) properties. Molecular dynamics simulations results for the ART1 protein demonstrated high stability over time. Additionally, resistomycin, borrelidin, tetracycline, and oxytetracycline were identified as the top-ranking ligands, exhibiting binding energies between −8.9 kcal/mol and −9.3 kcal/mol. These ligands exhibited favorable pharmacophore profiles, drug-likeness, and ADMET properties, indicating their potential safety and efficacy in humans. In conclusion, the selected actinomycete-derived ligands show promise for further research and development as potential anticancer agents targeting ART1. Full article

Sensor-enabled digital twins (DTs) are reshaping precision dairy nutrition by seamlessly integrating real-time barn telemetry with advanced biophysical simulations in the cloud. Drawing insights from 122 peer-reviewed studies spanning 2010–2025, this systematic review reveals how DT architectures for dairy cattle are conceptualized, validated, and deployed. We introduce a novel five-dimensional classification framework—spanning application domain, modeling paradigms, computational topology, validation protocols, and implementation maturity—to provide a coherent comparative lens across diverse DT implementations. Hybrid edge–cloud architectures emerge as optimal solutions, with lightweight CNN-LSTM models embedded in collar or rumen-bolus microcontrollers achieving over 90% accuracy in recognizing feeding and rumination behaviors. Simultaneously, remote cloud systems harness mechanistic fermentation simulations and multi-objective genetic algorithms to optimize feed composition, minimize greenhouse gas emissions, and balance amino acid nutrition. Field-tested prototypes indicate significant agronomic benefits, including 15–20% enhancements in feed conversion efficiency and water use reductions of up to 40%. Nevertheless, critical challenges remain: effectively fusing heterogeneous sensor data amid high barn noise, ensuring millisecond-level synchronization across unreliable rural networks, and rigorously verifying AI-generated nutritional recommendations across varying genotypes, lactation phases, and climates. Overcoming these gaps necessitates integrating explainable AI with biologically grounded digestion models, federated learning protocols for data privacy, and standardized PRISMA-based validation approaches. The distilled implementation roadmap offers actionable guidelines for sensor selection, middleware integration, and model lifecycle management, enabling proactive rather than reactive dairy management—an essential leap toward climate-smart, welfare-oriented, and economically resilient dairy farming. Full article

Metastatic dissemination defines a complex phenomenon driven by genetic forces and, importantly, determined by interaction between cancer cells and the surrounding stroma. Although the biologic and immune reactions which characterize the process have been widely and extensively evaluated, fewer data are available regarding the mechanical and physical forces to which circulating neoplastic clones are exposed. It should be hypothesized that this interaction can be modified in case of concomitant pathologic conditions, such as chronic vasculopathy, which frequently occurs in lung cancer patients. We here aim at analyzing and discussing the complex interplay between lung malignant transformation and arteriopathy, mainly focusing on the immune–inflammatory systemic reaction. Notably—in most instances—smoking-related fixed airflow obstruction, including but not limited to COPD, frequently coexists and contributes to both tumor progression and vascular complications. Attention is paid mainly to the analysis of the role of immune checkpoint inhibitors and their interaction with triple bronchodilation and antiaggregants. Understanding the biomechanical and molecular dynamics of lung cancer progression in altered vascular territories has several translational implications in defining risk stratification and in surgical planning and therapeutic targeting. Moreover, computational modeling of the physical forces which regulate the transit and extravasation of metastatic clones in altered contexts could be of help in deciphering the whole process and in determining more effective blockade strategies. Full article

The prospects for long-distance space flights are becoming increasingly realistic, and one of the key factors for their implementation is the creation of sustainable systems for producing food on site. Therefore, the aim of our work is to assess the prospects for using millet in biological life support systems and to create predictive models of yield components for automating plant cultivation control. The study found that stress from hypergravity (800 g, 1200 g, 2000 g, and 3000 g) in the early stages of millet germination does not affect seedlings or yield. In a closed system, millet yield reached 0.31 kg/m², the weight of 1000 seeds was 8.61 g, and the yield index was 0.06. The paper describes 40 quantitative traits, including six leaf and trichome traits and nine grain traits from the lower, middle and upper parts of the inflorescence. The compiled predictive regression equations allow predicting the accumulation of biomass in seedlings on the 10th and 20th days of cultivation, as well as the weight of 1000 seeds, the number of productive inflorescences, the total above-ground mass, and the number and weight of grains per plant. These equations open up opportunities for the development of computer vision and high-speed plant phenotyping programs that will allow automatic correction of the plant cultivation process and modeling of the required yield. Predicting biomass yield will also be useful in assessing the load on the waste-free processing system for plant waste at planetary stations. Full article