The article discusses the effect of substituting calcium ions Ca 2+ in certain positions of hydroxyapatite with iron ions Fe 2+ and Fe 3+ . It shows how to use molecular dynamics simulations with a skewed cell in periodic boundary conditions to find the density of hydroxyapatite. The obtained density values are in good agreement with known data and the results of density functional theory calculations. The article also describes the process of ion evaporation, melting, and sublimation of different hydroxyapatite samples. The molecular dynamics simulation results show significant differences in the temperatures of structural rearrangements for hydroxyapatite with Fe 2+ and Fe 3+ substitutions. For example, the hydroxyapatite sample with Fe 2+ ions experiences significant structural rearrangements in the temperature range of 727–827 °C, which is consistent with the experimental data on the decomposition of cation-substituted hydroxyapatite with iron ions at temperatures around 800 °C.
In the work presented, we study the nature of motion and distribution of charge along a chain during its Bloch oscillations in a constant electric field. Previous calculations have shown that the nature of the charge motion and distribution along the chain depends on the entire set of system parameters. For example, if only the intensity of an electric field or the initial charge distribution is changed, there may be either uniform motion or oscillatory motion when all other parameters of the system remain unchanged. In this work, the modeling of charge dynamics in a constant electric field was carried out in polynucleotide chains with different parameters of the oscillation frequency of chain sites and the friction coefficient. The conducted researches showed significant differences in the velocity and nature of charge motion in chains with different values of the specified parameters. Obviously, the study of charge motion in polynucleotide chains, whose parameters are close to those of DNA chains, is of the greatest interest. But numerical modeling of the charge dynamics in such chains tends to be more costly for computational resources. Therefore, in order to speed up the computations carried out, modelling is first performed in chains with more convenient parameters. It has been shown that in chains with parameters close to those of DNA chains, the charge can only move slightly along the chain, oscillating in the region of its initial position. With model values of the oscillation frequency of the chain sites and the friction coefficient, which significantly exceed the corresponding values of the parameters of DNA chains, the charge can move along the chain over a sufficiently large distance.
Transition metal dichalcogenide (TMD) monolayers are very promising for many applications, as well as graphene and hexagonal nitride boron layers (h-BN), especially in the fields of optics as emitters and detectors, in electronics as transistors. It is first of all due to the fact that they have a direct band gap Eg, which is dependent on external applied electric fields. To create such an electric field, it is proposed to use the field induced polarization of ferroelectric polymer layer such as polyvinylidene fluoride (PVDF) and poly(vinylidene fluoride-trifluoroethylene) P(VDF-TrFE). These polymers in the ferroelectric phase can create significant polarization in very thin layers, about 5 Å. By combining such ferroelectric layers and TMD layers, hybrid nanostructures can be created, that are convenient for design of new photodetectors with controlled properties. The outstanding properties of this hybrid structure are due, in particular, to the ultra-high electric field induced by ferroelectric polarization in PVDF or P(VDF-TrFE), which acts on the TMD layers and regulates the band gap Eg. In this work, we simulate such a hybrid structure based on PVDF and MoTe2, h-BN, graphene layers and study their features and properties. To calculate all the main properties, density functional theory (DFT) methods implemented in the Quantum ESPRESSO program are used. Semi-empirical PM3 method implemented in the HyperChem software package are used to model and study both individual layers of the hybrid structure and the features of their joint interaction, as well as to study vibration modes. The results obtained convincingly demonstrate a strong influence on the width of the band gap Eg and vibration modes frequencies of all the studied nanostructures due to the influence of the ferroelectric polarization P and the polarization-induced electric field E, which affects all the layers of the hybrid heterostructure.
Advanced fibrosis defines mortality risk in chronic liver disease. Current diagnostics fail to capture the molecular transition from early injury to permanent scarring. They often rely on binary classifications obscuring the biological drivers of disease progression. We mapped the gene expression profiles differentiating advanced (F3-F4) from early (F0-F2) fibrosis to determine their cellular origin and protein-level translatability across distinct disease etiologies. We analyzed public bulk RNA-sequencing data (E-MTAB-6863) to isolate genes distinguishing early from advanced fibrosis. We then deconvoluted these signals using single-nucleus RNA-sequencing (snRNA-seq; GSE179548) to identify specific cell types. We validated candidates proteomically in a multi-omics hepatocellular carcinoma dataset comparing viral versus metabolic etiologies. We identified a 105-gene signature associated with advanced fibrosis enriched for extracellular matrix organization and epithelial development. snRNA-seq localized this signal primarily to cholangiocytes rather than hepatic stellate cells indicating a ductular reaction. Proteomic validation showed a divergence. Structural markers (COL1A1, THY1) remained elevated regardless of etiology. Metabolic and ductular markers (AKR1B10, KRT23, SPP1) were specific to metabolic conditions (NAFLD alcoholic liver disease). Advanced fibrosis is fueled by a biliary epithelial response not just matrix deposition. The fibrotic scaffold is conserved across patients. The accompanying metabolic adaptations are etiology dependent. This suggests that effective biomarkers must be tailored to the specific metabolic context of the disease rather than relying on a universal fibrosis panel.
We present a novel framework for synthetic population generation using conditional Generative Adversarial Networks (cGANs) to infer individual-level health microdata from population-level marginals. Specifically, our approach synthesizes realistic microdata for target populations where only aggregate statistics are available, such as age-sex distributions and disease prevalence. To address the challenges of data scarcity and privacy constraints, we develop a cGAN architecture that captures and transfers complex intervariable relationships (e.g., age-disease and disease-disease dependencies) from a source population with microdata to a target population described by marginals. A hybrid loss function enforces the fidelity to the target marginals while preserving the epidemiological realism of the generated samples. We evaluated our method using the 2023 Behavioral Risk Factor Surveillance System (BRFSS) dataset from the USA, demonstrating strong alignment with real-world distributions across multiple states. The model accurately replicates disease co-occurrence patterns and age-disease correlations, even though these were not part of the conditioning data. Our results suggest that this method can enable scalable and privacy-preserving synthetic data generation, with promising applications in public health modeling and agent-based simulation, particularly in regions lacking detailed individual-level data.
Under natural conditions, plants are exposed to a complex set of adverse environmental factors that alter their RNA expression profiles. To date, RNA-Seq studies in Quercus robur L. have typically examined responses to individual stressors, and no meta-analysis integrating multiple conditions has been performed. Therefore, the aim of our study was to conduct a meta-analysis of RNA-Seq gene expression datasets associated with drought, powdery mildew infection, and insect herbivory in order to identify differentially expressed genes (DEGs), including transcription factors (TFs). We identified four overlapping DEGs responsive to all examined stressors: histone H4, flavonol synthase, acyltransferase GLAUCE-like, and an uncharacterized gene containing a DUF1223 domain. Analysis of transcription factor profiles using the iTAK v1.6 revealed 225 TFs in total, of which 159 were associated with insect feeding, 24 with powdery mildew infection, and 42 with drought stress. Four TFs were jointly downregulated under drought and insect herbivory. Under stress caused by powdery mildew and insects, nine TFs showed altered expression, including three WRKY genes and one representative each from the AP2/ERF, NAC, MYB, and GRAS families, while the remaining two belonged to the MYB-related family. No overlap in TF expression was observed between drought stress and powdery mildew. Weighted gene co-expression network analysis (WGCNA) identified 14 modules, six of which showed the strongest association with the examined traits and were selected for hub-genes detection. In total, six hub genes were identified; three of them were automatically annotated as NAC, GDSL, and CESA3, while the remaining genes remained uncharacterized. Overall, the meta-analysis demonstrates that plant responses to diverse stressors are mediated by both stress-specific and shared molecular mechanisms. Co-expression modules and transcription factors play a central role in these responses, enabling the identification of promising candidate genes for future functional studies and breeding efforts English oak.
One of the current directions in cell bioenergetics is a study of the molecular mechanisms underlying effective proton transport in proteins and surface proton-conducting structures along biomembranes. In this paper, we developed a mathematical model of lateral proton transport along a water-membrane interface. We considered a model of the proton movement in quasi 1D lipid domain structures in multicomponent lipid membranes, which are proposed to play proton-conducting lateral structures. The developed approach is based on a model of the three-component system including the subsystem of hydrogen bonded chains of surface water molecules interacting with the polar groups and hydrocarbon chains subsystems of lipid molecules. It was shown that collective soliton-type excitations can arise in this ternary system that lead to lateral proton transport accompanied by the movement of solitons, representing localized regions of compression in the subsystem of lipid polar groups and structural defects in the subsystem of hydrocarbon chains of lipid molecules. The results of modelling revealed that lateral proton transport in this coupled system is independent of the electric field along the membranes and can be determined by membrane curvature and elastic stress generated by proteins and ion channels embedded in biomembranes.
Abstract Sex-related differences in the aging of the human brain were studied using large array of experimental data. The open archive CamCan was used as a source of data: the magnetic encephalograms, co-registered with magnetic resonance images of the head, were obtained for each of 434 subjects (ages 18-87 years, mean age 54.7 ±18.4): 217 females (ages 18-87 years, mean age 54.5 ±18.4) and 217 males (ages 18-84 years, mean age 54.8 ±18.3). Recordings were split in 10-year age cohorts, each cohort consisted of equal number of men and women to calculate average intersex characteristics correctly. By massively solving the inverse problem, functional tomograms were calculated - the spatial distribution of elementary spectral components. Physiological noise was eliminated by joint analysis of MEG-based functional tomogram and magnetic resonance image for each subject. Then multichannel spectra were transformed into time series of the power of elementary current dipoles. Summary electric powers were calculated in six conventional frequency bands (1-4 Hz – delta; 4-8 Hz – theta; 8-13 Hz – alpha; 13-21 Hz – beta1; 21-30 Hz – beta2; 30-48 Hz – gamma), and sex differences in age-related changes were examined. It was found that in the youngest age cohort (18-29 years) the summary electrical power of the brain for males is 1.5 times greater than such power for females. For adults (30-69 years), male and female powers are approximately equal, while in older cohorts (70-87 years), male total brain power is greater. Age dependencies in various frequency bands are generally different for men and women, excluding higher frequencies 21-48 Hz. Basic conclusion can be made that after intersex averaging total electric power of the human brain is invariant through the lifespan from 18 to 87 years. The proposed method of joint MEG and MRI analysis can be used for further study of the sex-related details of brain sources in their connection with age changes.
Dengue, one of the most prevalent mosquito-borne viral infectious diseases of the twenty-first century, affects approximately 100–400 million individuals worldwide. The lack of definitive treatment has led to several in vitro and in silico investigations for potential antivirals. The RNA helicase of the dengue virus (DENV) is a vital protein that is important in viral replication. This study analyzed the antiviral activity of the flavanone naringenin and its derivatives through in silico and in vitro methods, targeting the DENV NS3 helicase enzyme. In this study, in silico screening based on molecular docking and ADMET properties of 20 derivatives of naringenin identified molecules such as 6-fluoro naringenin as the suitable compound. Further, comparative MD simulations and an MM-PBSA study for complexes like NS3-naringenin, NS3-6-fluoronaringenin, and NS3-ST-610 (standard inhibitor molecule) identified that naringenin was able to form a stable complex with the NS3 helicase enzyme of DENV2. These findings of the present study predict the promising anti-dengue nature of the compound naringenin as a replication inhibitor.
An original approach to coronavirus classification is proposed which is basing on presentation of gene analyzed (N-gene of nucleocapsid protein) by corresponding vector of amino acid codon frequencies and its subsequent comparison with vector of averaged codon frequencies for the known N-genes of viral taxon (one of the four coronavirus genera). Principal component analysis is used in non-standard way to determine whether frequency vector analyzed belongs to one of the taxons under consideration. Method was tested on 5769 N-genes of the four coronavirus genera and showed reliability of genus recognition above 95 %. Approach proposed for classification of the coronaviruses allows reducing dimension of codon frequency vector to 28 components without decrease of reliability, by considering the most significant amino acid codon frequencies in N-gene. The approach refers to alignment free methods which become increasingly popular in the last decade for virus classification.
The theta rhythm synchronizes neural activity in the processes of attention and memory. However, the mechanisms of synchronization of neural activity during theta rhythm generation are not known. We propose a mathematical model explaining the distribution of the CA1 field neurons over the theta rhythm phase. We examined a network consisting of 10 types of inhibitory cells: (PV) parvalbumin and (CCK) cholecystokinin basket cells, axo-axonal, bistratified, neurogliaform, perforant path-associated, interneuronal-specific (subtypes R-O and RO-O), Ivy and OLM neurons. The network received four excitatory inputs from the CA3 field, the medial entorhinal cortex, and two types of local pyramidal neurons of the CA1 field. We have shown that it is possible to fit the parameters of connections in the model that neurons form experimentally observed phase relationships relative to the theta rhythm. For most types of neurons, excitatory inputs add up and give a maximum near the peak of discharges in the theta cycle. The peak of inhibitory inputs falls on the opposite phase of the theta rhythm, due to this, activity slows down from the opposite phase of the theta rhythm. The model steadily reproduces the phase relations over the entire frequency range of the theta rhythm for most types of interneurons.
The main approaches to studying the human heart using non-invasive measurements are discussed. Methods such as echocardiography, computed tomography, and magnetic resonance imaging are used to study the heart's structure. Cardiac function is studied primarily using electrocardiography and magnetic cardiography. Data sets available for use in mathematical modeling of the heart to study its function and for diagnostic purposes are described. A method of functional tomography is proposed that transforms a set of time series into a spatial distribution of electrical or magnetic sources. Experimental magnetic cardiography data obtained at the Center for Neuromagnetism, New York University, at the Kurchatov Institute National Research Center and the Kotelnikov Institute of Radio Engineering and Electronics of the Russian Academy of Sciences, were used to reconstruct the three-dimensional functional structure of the heart and time series of electrical activity of the heart. The current dipole found from the magnetic cardiogram was successfully localized in the space of the reconstructed heart structure, and its amplitude and direction changed in accordance with the phase of the cardiac cycle. The method can be used for a detailed study of the spatial distribution of elementary sources of electrical activity of the human heart based on multichannel non-invasive measurements.
Internet resources providing an open access to experimental functional brain data, including electric and magnetic encephalography, structural and functional MRI data obtained in a number of research projects studying the human brain are reviewed. The usage of such resources is beneficial for the development and adjustment of new methods for processing and analyzing encephalograms, as well as for verifying and comparing results.
The accuracy of resolution estimation of an experimental cryo-EM scattering density map has a decisive effect on the accuracy of determining the uncertainty parameters of atomic positions (ADP parameters), when refined in direct space. The previously developed methodology for local resolution estimation of cryo-EM maps using direct space methods was applied to the study of maps for type I pilus from E. coli. A study of four maps obtained by three groups of researchers showed that, for each of the three standard maps, the resolution can be considered constant and is about 2.6–2.7 Å, while for the sharpened map, the resolution is higher and is equal to 2.3 Å.
Within creation of the mathematical model to describe the human respiratory system, we accomplished numeric investigation of non-stationary dust-containing airflow as well as dust particle deposition in the lower airways with the real anatomic geometry based on CT scans. Inhaled air is considered a multi-phase mixture of a homogenous gas and solid dust particles. Motion of a basic carrier gas phase is described using the Euler approach. Solid dust particles are a dispersed carried phase, which is described with the Lagrange approach. The k-ω model is used to describe turbulence. We consider non-stationary airflow during calm inhalation. The article presents calculated flow streamlines for the velocity of particles in inhaled air in the lower airways at different moments. We quantified a share of deposited particles (SDP) with various dispersed structure (between 10 nm and 100 µm) and density (1000 kg/m3, 2000 kg/m3, 2700 kg/m3) in the lower airways; the article provides computed motion paths of particulate matter. Solid particle deposition in the airways has different efficiency depending on particle sizes and density. SDP goes down as their sizes and masses decrease. Particle density mostly influences differences in deposition of micro-sized particles (2.5–20 µm): as particle mass and density grow, SDP in the airways also increases. SDP with their diameter being less than 1 µm amounts to approximately 20 % of all the particles that reach the inlet to the trachea. According to the results obtained by numeric modeling, the greatest share of dust particles penetrates the right main bronchus, predominantly the right middle and inferior lobar bronchi. Dust particles are able to induce diseases of the lungs, pneumoconiosis included.
This work is the final part of a series of studies devoted to the creation of a hierarchy of basic models of the biokinetics of aseptic inflammation, which represents a mathematical formalization of fundamental general biological concepts about the participation of components of the innate immune system in protective and adaptive reactions in the focus of inflammation. The idea of local spatial uniformity of the modelled process is accepted. The formation of a hierarchy of models is based on the bottom-up principle, which allows you to first focus on a variety of details, and then move on to representing the system in its most general form, i.e. at the level of «major degrees of freedom». The idea of reduction is based solely on general biological considerations reflecting objective data on the role of components of the innate immune system in the functioning of the protective and adaptive mechanism of inflammation. The sequential reduction resulted in eight basic mathematical models of the biokinetics of inflammation. Each of them has successfully passed the validation procedure using experimental time series that characterize the dynamics of in vivo inflammation in excisional skin wounds in mice and several other living models. It is also shown that the proposed structure of the equations of all models makes it possible to adequately reproduce the functioning of the fundamental mechanism of M1/M2 polarization and reprogramming of macrophages. The robustness of each model is confirmed by an analysis of sensitivity to small perturbations of parameters and a cycle of «diagnostic» checks, which consist in a qualitative assessment of possible changes in the main scenario of an acute inflammatory reaction in the well-known disorders of the leukocyte formula or functions of blood cells. The study of the dependence on the initial conditions showed that all models have the property of multistability in a biologically significant range of parameter values and solutions of the obtained systems of differential equations, due to which the switching of the inflammation scenario from acute to chronic can be mathematically reproduced. An important place is assigned to the final stage of reduction, focused on defining the minimal model as a key constructive element, which is common to all mathematical models of the hierarchy and the «carrier» of the multistability property. It is shown how only a few «main» ones can be distinguished from the set of degrees of freedom (in the phase space of states), so that the final result was a biologically justified reduction in the dimension of the phase space of states by more than half while preserving the most significant properties of the most complete model and the real object. In the context of aseptic inflammation, the «main degrees of freedom» can be attributed both a priori and according to mathematical modelling to the quantitative characteristics of subpopulations of blood cells (active platelets) and the innate immune system (neutrophils, macrophages). The results of the reduction procedure indicate that the mathematical idealization adopted in the hierarchy of models supports the following fundamental biological fact: the self-organization of immune cells within the framework of a genetically determined inflammatory program, their adaptation in the microenvironment cannot effectively occur without the cytokine involvement. The practical significance of the proposed models lies in the fact that they are an effective self-sufficient tool for studying a genetically determined program of cellular and molecular immune response to damage of any etiology, and can also be used as a subsystem in complex multilevel multiphysical models of pathogenesis mechanisms of most human diseases. Applications of the developed mathematical models can include ischemic heart attacks, neurodegenerative and oncological processes, damage caused by mechanical or chemical factors, surgical interventions, i.e. the widest range of general pathological processes, in which the innate immune response is an important pathogenesis factor.
Between 1990 and 2022, the number of people with diabetes increased from 200 million to 830 million, with the majority of cases attributed to non-insulin-dependent type 2 diabetes. This condition often leads to severe complications such as blindness, kidney failure, myocardial infarction, stroke, and limb amputations. According to WHO estimates, more than 2 million people died in 2021 from diabetes and kidney-related diseases. Global healthcare expenditures related to diabetes were estimated at 966 billion USD in 2021 and are projected to reach 1.054 trillion USD by 2045. Early diagnosis and a personalized approach to treatment can slow disease progression and reduce the risk of vascular complications. At early stages, lifestyle modification alone can achieve long-term normoglycemia without the need for medication. However, nearly half of all people with diabetes worldwide remain undiagnosed. One of the key challenges remains the prevention of hypoglycemia, particularly nocturnal hypoglycemia, which can lead to dangerous complications. Today, the use of continuous glucose monitoring data combined with machine learning algorithms enables the prediction of hypoglycemia risk, taking into account individual patient characteristics. Modern artificial intelligence models rely on large-scale, multimodal data analysis and can adapt in real time. Addressing the issue of data imbalance improves the reliability of predictions. Further progress requires the development of large data libraries and secure systems for information exchange between monitoring devices and clinical applications. The implementation of such technologies enhances treatment effectiveness and improves patients' quality of life. This review highlights the prospects for integrating artificial intelligence into diabetes monitoring and prediction.
The new technology was developed to calculate spectral characteristics of various compartments of the human brain. This technology combines two types of spatial data: 1) MEG-based functional tomogram presenting spatial distribution of the electric sources and 2) anatomical structure of the brain estimated by the magnetic resonance imaging. Multichannel magnetoencephalograms were used to calculate the functional tomogram by the precise frequency-pattern analysis, decomposing brain activity into a set of elementary oscillations. In the functional tomogram, unique spatial location corresponds to each spectral component, generated by the equivalent current dipole. Based on this fact, we calculated partial spectra – sets of frequencies, generated by various anatomical regions of the brain. Individual spatial structure of the brain compartments was determined by the annotated segmentation of magnetic resonance tomogram for each subject under study. The partial spectrum of the brain compartment was composed of the set of frequencies, localized into this compartment. Here we calculated partial spectra of 15 compartments of the brain for 600 subjects, using MEG and MRI datasets obtained from the open archive CamCAN. Average spectra were calculated and the distribution of the spectral power between brain compartments was estimated.
The work is devoted to the study of the dependence chlorophyll a and P 700 ratio (Chl a / P 700 ) on light intensity. Literature data on the structure of pigment-protein supercomplexes, mechanisms of redistribution of excess light energy are reviewed. The literature analysis showed that Chl a / P 700 ratio in the culture of lower photoautotrophic organisms determines the specific growth rate or photosynthesis rate. Chl a / P 700 value of can be used as indicator of the microalgae culture physiological state. The basic principles of the photobiosynthesis modeling were used for described the light influence on Chl a / P 700 value. The division of the whole chlorophyll pool into two conventional groups is proposed. Structural chlorophyll is associated with the nuclear part of the photosystems. Antenna chlorophyll is linked with peripheral proteins of the light-harvesting complex. The ratio of all biomass structural components is considered to be unchanged. The antenna chlorophyll content is determined by the rate of its synthesis from reserve biomass pool and light-dependent degradation. The system of equations is solved for the stationary state of turbidostat. It is shown that the ratio of antenna chlorophyll and structural biomass concentrations determines Chl a / P 700 value. The relationship between this parameter and light intensity was established. The verification of the model was carried out for the turbidostat Arthrospira platensis culture. Three ranges of light intensity were identified, which are characterized by different photoadaptation mechanisms: the area of antenna chlorophyll loss at low intensities; the area of antenna chlorophyll accumulation; the area of metabolic limitation of antenna protein biosynthesis with the participation of hydrogen peroxide. The theoretical basis for this division is provided by experimental results that describe the different mechanisms of light energy excess distribution at low and high light intensities. For each area an analytical expression which describing the Chl a / P 700 light kinetics was obtained. Comparison of theoretical curves and experimental data for A. platensis culture allowed us to determine species-specific model coefficients, which remained unchanged at all three curve sections. The obtained results allow us to develop mathematical algorithms for regulating the pigments content in the microalgae biomass.
We have developed algorithm MinSufPref for computing of the probability (P-value) of S occurrences of words from a set Η in a random sequence of length N. It is widely used in computer sciences as a criterion for selecting of patterns of a special kind. In particular, in bioinformatics, when searching for functional fragments in biological sequences. The proposed algorithm is based on the previously developed algorithm SufPref for exact P-value computation. SufPref uses an overlap graph for calculations, the nodes of the graph correspond to the words from Η and their overlaps. The restrictions of SufPref are that it admits only sets consisting of words of a same length, and also the algorithm may not be applicable for sets consisting of a large number of words. In this paper, the algorithm is expanded to sets consisting of words of arbitrary lengths, but with the constraint that a set can not contain words occurring in other its words. Also the R-equivalence relation on the prefixes of pattern words is introduced, the overlap graph is minimized in accordance to this relation. The space and time complexities of the main part of MinSufPref are linear on the number of nodes and edges of the minimized overlap graph. The time complexity is independent of the alphabet size and the lengths of words in Η. We have performed computer experiments that showed high efficiency of MinSufPref compared to similar algorithms.
