Browse Articles

Uncovering functional connectivity patterns predictive of cognition in youth using interpretable predictive modeling

Brain-wide association studies using functional MRI have advanced our understanding of how behavioral traits relate to individual variability in brain function. These studies typically identify functional connectivity (FC) patterns linked to behavioral traits using either whole-brain or region-wise predictive models. However, whole-brain models often struggle with generalizability and interpretability due to the high dimensionality of FC data, while region-wise models isolate predictions, limiting their ability to capture the integrated contributions of brain-wide FC patterns. In this study, we introduce an interpretable predictive model that learns fine-grained FC patterns predictive of behavioral traits, jointly at the regional and participant levels, to characterize the overall association of FC patterns with a target trait. Our model jointly learns a relevance score and a dedicated prediction function for each brain region, then integrates the regional predictions using the relevance scores as weights to generate a participant-level prediction, capturing the collective association of FC patterns with the trait. We validated our method using FC data from 6,798 participants in the Adolescent Brain and Cognitive Development (ABCD) study to predict cognition. Our model identified the cingulo-parietal, retrosplenial-temporal, dorsal attention, and cingulo-opercular networks as collectively predictive of cognitive traits, achieved competitive prediction accuracy, and enabled detailed characterization of fine-grained FC differences across cognitive domains. The learned relevance scores enhanced region-wise predictions of longitudinal cognitive measures in the ABCD cohort and cognitive traits in the Human Connectome Project Development cohort. These findings suggest that our method effectively characterizes generalizable and fine-grained FC patterns linked to cognition in youth.

Assessment of potentially toxic elements in agricultural and urban soils of lesser Poland using ICP-MS and uncertainty estimation

Evaluation of biomass and vegetative characteristics of mesquite (Prosopis juliflora) afforestation in arid area of Iran

A protein dynamics–based deep learning model enhances predictions of fitness and epistasis

Deep learning has advanced our ability to assess the effects that individual mutations have on protein function; however, predicting the complex interplay between two or more mutations remains challenging. Here, we seek to address this challenge by building a deep learning framework that incorporates information related to protein dynamics. Namely, we build a neural network architecture using a physics-based metric called the Asymmetric Dynamic Coupling Index (DCI asym ), which quantifies the degree to which each member of a pair of residues influences the flexibility of the other. DCI asym enables us to train models through an allosteric Graph Neural Network (GNN) in which each residue is linked to its distant dynamic influencers. Despite not being trained on experimental epistasis data, our GNN consistently outperforms existing approaches on deep mutational scanning datasets across four distinct proteins, highlighting its enhanced capacity to model epistatic interactions. Our GNN model was then challenged to predict the functions of 37 novel, computationally designed TEM-1 β-lactamase variants of unknown function, and it demonstrated excellent predictive accuracy for these variants. Thus, our GNN provides a pathway for better assessing the impact of multiple mutations on protein function, including epistatic relationships and mutations that have profound effects on activity despite being spatially far from the active site.

Enhancing indoor monitoring of visually impaired people using temporal convolutional network with optimization model in IoT environment

An interpretable machine learning approach based on SHAP, Sobol and LIME values for precise estimation of daily soybean crop coefficients

Abstract Increasing water scarcity and climate variability have intensified the need for precise agricultural irrigation management. Accurate estimation of crop coefficients (Kc) is critical for determining crop water requirements, especially in arid and semi-arid regions. However, conventional methods for estimating Kc often rely on generalized plant characteristics, which may not account for local climatic variations. In this study, we address this challenge by predicting the daily crop coefficient for soybean using four machine learning models: Extreme Gradient Boosting (XGBoost), Extra Tree (ET), Random Forest (RF), and CatBoost. These models were trained on meteorological data from Suhaj Governorate, Egypt, spanning 1979–2014. Additionally, SHapley Additive exPlanations (SHAP), Sobol sensitivity analysis, and Local Interpretable Model-agnostic Explanations (LIME) were applied to evaluate model interpretability and consistency with physical processes. Among the models evaluated, the ET model achieved the highest accuracy, with r = 0.96, NSE = 0.93, RMSE = 0.05, and MAE = 0.02. XGBoost and RF also performed well, each obtaining r = 0.96, NSE = 0.92, RMSE = 0.06, and MAE = 0.02. In comparison, CatBoost demonstrated slightly lower accuracy, with r = 0.95, NSE = 0.91, RMSE = 0.06, and MAE = 0.02. SHAP and Sobol analyses consistently identified the antecedent crop coefficient [ $$\:Kc(d-1)$$ ] and solar radiation (Sin) as the most influential variables. LIME results revealed localized variations in predictions, reflecting dynamic crop-climate interactions. This study underscores the importance of integrating interpretable machine learning models to enhance both predictive accuracy and reliability while maintaining alignment with critical physical processes. The proposed framework offers a robust tool for improving daily Kc estimation, thereby supporting more sustainable irrigation practices and climate-resilient agriculture.

Integrative mapping reveals molecular features underlying the mechanism of nucleocytoplasmic transport

Nuclear pore complexes (NPCs) enable rapid, selective, and robust nucleocytoplasmic transport. To explain how transport emerges from the system components and their interactions, we used experimental data and theoretical information to construct an integrative Brownian dynamics model of transport through an NPC, coupled to a kinetic model of transport in the cell. The model recapitulates key aspects of transport for a wide range of molecular cargoes, including preribosomes and viral capsids. Our model quantifies how flexible phenylalanine-glycine (FG) repeat proteins create an entropic barrier to passive diffusion and how this barrier is selectively lowered in facilitated diffusion by the many transient interactions of nuclear transport receptors with the FG repeats. Selective transport is enhanced by “fuzzy” multivalent interactions, redundant FG repeat mass, coupling to the energy-dependent RanGTP concentration gradient, and exponential dependence of transport kinetics on the transport barrier. Our model will facilitate rational modulation of the NPC and its artificial mimics.

A deep learning framework with hybrid stacked sparse autoencoder for type 2 diabetes prediction

A novel adaptive transformer based quantum intrusion detection system for software defined networks

Temporal and spatial coordination of DNA segregation and cell division in an archaeon

Cells must coordinate DNA segregation with cytokinesis to ensure that each daughter cell inherits a complete genome. Here, we explore how DNA segregation and division are mechanistically coupled in archaeal relatives of eukaryotes, which lack Cyclin-dependent kinase (CDK)/Cyclins. Using live cell imaging, we first describe the series of sequential changes in DNA organization that accompany cell division in Sulfolobus, which computational modeling shows likely aid genome segregation. Through a perturbation analysis we identify a regulatory checkpoint which ensures that the compaction of the genome into two spatially segregated nucleoids only occurs once cells have assembled a division ring—which also defines the axis of DNA segregation. Finally, we show that DNA compaction and segregation depend, in part, on a ParA homologue, SegA, and its partner SegB, whose absence leads to bridging DNA. Taken together, these data show how regulatory checkpoints like those operating in eukaryotes aid high-fidelity division in an archaeon.

The prevalence, recognition, and treatment of depression and anxiety symptoms among Chinese cardiovascular outpatients

Investigation of Akt1 behavior on gold surface from molecular dynamics insight

Abstract In the past few years, gold nanoparticles (AuNPs) have shown great roles in biomedical areas. They can interact with proteins and change their structure and function. The serine/threonine kinase AKT plays a key role in cellular processes. Therefore, the AKT protein is known as a drug target for cancer treatment. In this study, we assessed the effect of gold nanoparticles on the AKT1 protein using molecular docking and molecular dynamics simulation. The results show that the AKT1 protein binds to citrate-coated gold surface predominantly through electrostatic and hydrophobic interactions. The RMSF calculations show that the AKT1 protein in the presence of gold nanoparticles exhibits less dynamic than the free state. The presence of gold nanoparticles causes the protein to have less compactness. The linker domain in the inactive conformation, and the regulatory domain and the glycine-rich loop in the active conformation of the AKT1 protein have higher dynamics than other regions. Furthermore, free energy landscape calculations show that AKT1 protein has a more conformational entropy in complex states in two active and inactive conformations. The results show that gold nanoparticles can affect the AKT1 protein and as a result, inhibit the phosphorylation flow in the protein signaling pathway in cancer cells.

Formation of a complex between TMEM217 and the sodium-proton exchanger SLC9C1 is crucial for mouse sperm motility and male fertility

Sperm motility is essential for male fertility and is tightly controlled by signaling events in the flagellum. Slc9c1 encodes a sperm-specific Na + /H + exchanger (sNHE/SLC9C1) that localizes to the flagellum and is indispensable for sperm motility and male fertility. SLC9C1 is unique among Na + /H + exchangers in that it possesses a voltage-sensing domain (VSD), the physiological function of which remains poorly understood in mammals. Here, by analyzing coevolving genes with Slc9c1 , we identified Tmem217 , which encodes a transmembrane protein that is localized in the sperm flagellum. Knockout (KO) of Tmem217 in mice resulted in sperm motility defects and male infertility, phenocopying Slc9c1 KO mice. Coimmunoprecipitation and structural prediction analyses indicated that TMEM217 binds to SLC9C1 via its VSD. Further analyses indicated that the amounts of SLC9C1 and its associated protein, soluble adenylyl cyclase (sAC), were lost in mature Tmem217 KO spermatozoa, leading to disrupted 3′,5′-cyclic monophosphate (cAMP) signaling pathways. Remarkably, cAMP analogs restored the impaired motility and fertilizing ability of Tmem217 KO spermatozoa in vitro, validating the essential role of TMEM217 in regulating cAMP production. Our findings indicate that the association of TMEM217 with SLC9C1 via its VSD is critical for the proper organization and function of the SLC9C1–sAC–cAMP axis in mature spermatozoa.

Central composite design optimized fluorescent method using dual doped graphene quantum dots for lacosamide determination in biological samples

A life cycle assessment of disposing intra-operative collected fluids, a comparative study between the Neptune 3 versus canister drainage

Abstract The Neptune 3 drainage system is developed as an alternative to canister systems for collecting surgical fluids in the operating room. This study investigates the difference in the environmental impact between Neptune 3 and canister systems, evaluating all individual stages of the product life cycle of the Neptune and canisters. Using the RECIPE model, 17 impact categories (midpoints) were defined and results were aggregated into three endpoint categories (human health, ecosystems, resources). The volume of waste was varied in a setup in a hospital in the Netherlands and included different fluid volume collection scenarios performed over seven years: high-volume (2.0–24 L waste) and low-volume (0.1–0.5 L waste). In both high and low volume procedures, Neptune 3 has a lower environmental impact compared to canisters for global warming (15–89% reduction), ozone formation terrestrial ecosystems & human health (24–91% reduction), fossil resource scarcity (36–92% reduction) and water consumption (44–106% reduction). In high volume scenarios (5 + Liters) Neptune also has a lower impact in stratospheric ozone depletion, fine particulate matter formation, terrestrial acidification for the high volume scenarios (5 L or more). In the case of ionizing radiation, freshwater eutrophication, and human carcinogenic toxicity the Neptune has a lower impact only in the very large volume procedures (10 + Liter). By aggregating the mid-point results to end-point results, it is observed that the Neptune system is beneficial for resources in each scenario, and for human health and ecosystems for procedures with larger volumes. Results from this LCA demonstrate that the Neptune 3 system is environmentally beneficial compared to canisters. This study provides valuable information for policymakers and hospital decisionmakers to treat their surgical waste in an environmentally friendly way.

Information sampling and Bayesian belief formation in statistical judgment

The statistical properties of data are often communicated using visual graphs, like scatterplots. However, decision makers make systematic errors when processing these graphs, with important consequences for statistical communication in science, medicine, and policy. We propose that decision makers are Bayesian learners, who learn optimally given the data points that they attend to. Accordingly, judgment errors arise from biased sampling of information from graphs. We tested our theory in four eye-tracking experiments (total N = 421), in which participants made correlation judgments from scatterplots of both experimentally manipulated data (Experiment 1) and real data (Experiment 2), as well as plots with different display formats (Experiments 3 and 4). Participants’ judgments displayed several known biases, including underestimation of absolute correlations and sensitivity to irrelevant visual features. Importantly, the (optimal) Bayesian belief updating model, trained on the sensory inputs from visual information search, predicted both participants’ judgments and associated biases with high accuracy in all the experiments. Additionally, a computational model of participants’ information sampling processes, combined with the Bayesian model, reproduced all behavioral regularities. These results shed light on the cognitive mechanisms of belief formation, show how statistical judgments can be quantitatively predicted and manipulated, and provide insights for data visualization and statistical communication.

Decoding the mind of self-compassion through a topic modeling analysis of 9000+ free-text narratives

Assessment of environmental sustainability of solid waste management system of Dhaka city through life cycle analysis

FETCH enables fluorescent labeling of membrane proteins in vivo with spatiotemporal control in <i>Drosophila</i>

Fluorescent labeling approaches are crucial for elucidating protein function and dynamics. While robust methods to monitor gene transcription are widespread, the visualization of proteins in vivo is more elusive. To meet this challenge, we developed Fluorescent Endogenous Tagging with a Covalent Hook (FETCH) to label cell surface proteins (CSPs) in vivo through a stable covalent bond mediated by the DogTag-DogCatcher peptide partner system. FETCH leverages a spontaneous covalent isopeptide bond that forms between the 23-amino acid DogTag and the 15-kDa DogCatcher. Unlike most tags that work best at protein termini, DogTag functions well in protein loops, expanding the range of sites that can be targeted in proteins. In FETCH, DogTag is introduced into extracellular loops of CSPs through genome engineering, enabling covalent bond formation with a genetically encoded DogCatcher-GFP fusion protein that can be secreted from a variety of cell types in intact animals. To identify optimal DogTag insertions into CSPs, we describe a flow cytometry–based platform for rapidly screening candidates in vitro. We demonstrate the ability to tag and visualize three members of the immunoglobulin superfamily (IgSF) in vivo: the transmembrane protein mCD8 and two GPI-anchored proteins belonging to the DIP-Dpr interactome that interact biophysically to facilitate neuronal target recognition at Drosophila neuromuscular and brain synapses. FETCH enables precise temporal and spatial control to visualize tagged proteins in vivo, features that are adaptable to a multitude of applications for modifying any cell surface protein.

Investigation of multifaceted wound healing effect of exopolysaccharide (EPS) produced from probiotic strain Lactiplantibacillus plantarum GD2 as in vitro and in ovo

Abstract Skin wounds may threaten quality of life and cause serious complications. This study aimed to investigate the effects of lyophilized exopolysaccharide (L-EPS) obtained from the probiotic strain Lactiplantibacillus plantarum GD2 on various stages of wound healing. The results revealed that L-EPS accelerated in vitro wound healing and increased COL1A1 in L929 cells. L-EPS affected the TGF-β1/Smad signaling pathway by increasing the expression of the TGF-β1, Smad2, Smad3, and Smad4 genes. L-EPS also exerted anti-inflammatory effects by reducing the gene expression of IL-1β, IL-6 and iNOS in TNF-α-induced fibroblasts. Additionally, L-EPS demonstrated fibroproliferative effect on both healthy and TNF-α-induced fibroblasts. Furthermore, L-EPS was found to have a proangiogenic effect in ovo chorioallantoic membrane (CAM) model. This study presents the first-ever characterization of the multifaceted effects of L-EPS derived from the probiotic strain L. plantarum GD2 on wound healing. Our findings highlight the potential of L-EPS as effective agent for wound healing and suggest possible application in the development of wound healing biomaterials. By elucidating the mechanism of action of L-EPS in wound healing, this research may provide new perspectives for advanced treatment strategies in the field of wound care.