Biology subjects

Kurata, H.

Publications and source records attributed to Kurata, H..

11 recordsLinked to original sources

bioRxiv2026

RNArefine: AI-guided Atomic-Level Refinement of RNA Structures

Considerable progress has been made in AI-driven RNA structure prediction, but the resulting models often lack complete atomic details or suffer from severe stereochemical distortions and incorrect local interactions. We present RNArefine, an AI-guided hierarchical framework for atomic-level RNA structure refinement. RNArefine first predicts base-pairing and base-stacking interactions using geometric attention networks and then integrates the interactions with physics-based force fields to guide a two-step refinement strategy consisting of Monte Carlo conformational sampling followed by L-BFGS energy optimization. Large-scale benchmark experiments on both sequence-based prediction models and cryo-EM-derived structures demonstrated that RNArefine consistently improves stereochemical quality, interaction fidelity and physically penalized structural accuracy while preserving global topology. When applied to blind CASP16 RNA prediction models, RNArefine improved ranking scores for 28 of the top 30 groups. These results establish RNArefine as a robust open-source framework for transforming raw RNA folds into physically realistic atomic models for downstream structural and therapeutic applications.

bioengineering↗

bioRxiv2026

Computer experimentation on E. coli ammonium transport and assimilation reveals mechanisms for energy coupling, balanced futile cycling, and robust growth

Nitrogen is essential for all life forms, and microorganisms prefer ammonium as a nitrogen source. Due to the low affinity of glutamine synthetase (GS) for ammonium, E. coli must maintain high intracellular ammonium (NH4+) concentrations to sustain its rapid growth. Under ammonium limitation, E. coli imports ammonium through the transporter AmtB and incorporates it into glutamine by using GS. On the basis of structural and mutagenesis information, mechanisms have been proposed for the transport of ammonia (NH3) and protons by AmtB through spatially (partly) separate routes. These mechanisms do not explain the required coupling between proton and ammonia transports. How does the membrane potential push the ammonia inward so as to attain high concentrations near GS? We here compare six candidate kinetic models of E. coli ammonium transport and assimilation in terms of how they reproduce experimental data from the literature: three variants of the electro-binding model in which the membrane potential affects AmtB-NH4+ binding, and three variants of the electro-flipping model in which it influences the conformational flip of the transporter. The computer simulations decide that the electro-binding models are 28 times more plausible than the electro-flipping models and suggest that the transmembrane electric potential affects AmtB-NH4+ binding from the cytoplasmic side. The addition of kinetic and thermodynamic features to existing structural information plus our requirement of an explanation of the coupling, suggest a new spatiotemporal mechanism of coupling of ammonia and proton flows in AmtB. Further simulations show that GS and AmtB regulation is coordinated via both the uridylyltransferase/uridylyl-removing enzyme (UTase) and 2-oxoglutarate binding, allowing the cell to minimize futile cycling while maintaining rapid growth. The free energy cost of transport-related futile cycling exceeded that of the GS reaction itself. Moreover, AmtB enabled robust growth under varying ammonium concentrations and pH levels, albeit at a cost of futile cycling that became substantial at low ammonium. These findings highlight the crucial roles of GS and AmtB in E. colis adaptations and provide new insights into the trade-off mechanism between nutrient acquisition and energy efficiency.

systems biology↗

bioRxiv2025

Meta-PseU: A Meta-Classifier for Robust Prediction of RNA Pseudouridine Modification Sites from Long Sequences

Pseudouridine ({Psi}) represents one of the most abundant and evolutionarily conserved RNA modifications. {Psi} provides an additional hydrogen-bond donor that enhances RNA structural stability and modulates translation. It participates in diverse biological processes, including RNA-protein interactions, splicing, translational control, and stress responses. Aberrant pseudouridylation is implicated in cancer, neurodegenerative disorders, and autoimmune diseases. Despite its biological importance, experimental identification of {Psi} sites remains time-consuming and costly, limiting the feasibility of transcriptome-wide profiling. Computational approaches have therefore become essential complements to experimental techniques. However, current machine-learning and deep-learning predictors suffer from limitations such as small datasets and limited generalizability. To overcome these issues, we have constructed new long-sequence datasets derived from RMBase 3.0 and developed Meta-PseU, a logistic regression-based meta-classifier that stacks multiple single-feature or baseline classifiers across three species of human, mouse, and yeast. Meta-PseU substantially reduced the performance gaps between training and independent test datasets, presenting superior generalization. Meta-PseU substantially outperformed state-of-the-art predictors and achieved increasing accuracy with increasing sequence length. This work offers a new framework for robust {Psi}-site identification by using long sequences. The datasets and programs are freely accessible at https://github.com/kuratahiroyuki/MetaPseU.

bioinformatics↗

bioRxiv2025

GDA-Pred: Generative AI-Driven Data Augmentation for Improved Prediction of IL-6 and IL-13 Inducing Peptides

The identification of interleukin-6 (IL-6) and interleukin-13 (IL-13) inducing peptides is crucial for accelerating drug discovery targeting cancer, immune disorders, and infectious diseases. However, experimental screening methods remain time-consuming and costly. To address these limitations, various machine learning and deep learning models have been developed, yet their performance is still constrained by the limited availability of experimentally validated data. In this study, we propose a generative AI-driven data augmentation (GDA) framework and predictor, GDA-Pred, to improve the prediction performance of state-of-the-art (SOTA) classifiers for identifying IL-6 and IL-13 inducing peptides. GDA expands the training dataset by generating novel peptide sequences using three types of generative AI models: generative adversarial networks (GANs), diffusion models (DMs), and variational autoencoders (VAEs). The GDA framework is defined by four key parameters: the type of generative model, the sequence identity cutoff, the probability threshold (PT) for selecting generated peptides, and the augmentation ratio (AR) between generated and real peptides. Since optimizing these parameters is challenging with small datasets, we adopt a case study-oriented proof-of-concept approach using a moderately sized dataset of anti-inflammatory peptides (AIPs) to derive interpretable optimal settings. The performance of the optimized GDA was evaluated using stratified 5-fold cross-validation with cluster-based partitioning and a hold-out test on benchmark datasets. The optimized GDA was then applied to SOTA classifiers, collectively termed GDA-Pred, to identify IL-6 and IL-13 inducing peptides, both of which are limited by small dataset sizes. GDA-Pred substantially improved the prediction performance for both cytokine-inducing peptide tasks, demonstrating the feasibility of GDA-Pred as a robust and generalizable framework.

bioinformatics↗

bioRxiv2022

ICAN: interpretable cross-attention network for identifying drug and target protein interactions

Drug-target protein interaction (DTI) identification is fundamental for drug discovery and drug repositioning, because therapeutic drugs act on disease-causing proteins. However, the DTI identification process often requires expensive and time-consuming tasks, including biological experiments involving large numbers of candidate compounds. Thus, a variety of computation approaches have been developed. Of the many approaches available, chemo-genomics feature-based methods have attracted considerable attention. These methods compute the feature descriptors of drugs and proteins as the input data to train machine and deep learning models to enable accurate prediction of unknown DTIs. In addition, attention-based learning methods have been proposed to identify and interpret DTI mechanisms. However, improvements are needed for enhancing prediction performance and DTI mechanism elucidation. To address these problems, we developed an attention-based method designated the interpretable cross-attention network (ICAN), which predicts DTIs using the Simplified Molecular Input Line Entry System of drugs and amino acid sequences of target proteins. We optimized the attention mechanism architecture by exploring the cross-attention or self-attention, attention layer depth, and selection of the context matrixes from the attention mechanism. We found that a plain attention mechanism that decodes drug-related protein context features without any protein-related drug context features effectively achieved high performance. The ICAN outperformed state-of-the-art methods in several respects and revealed with statistical significance that some weighted sites in the cross-attention weight represent experimental binding sites, thus demonstrating the high interpretability of the results. Key pointsWe created the interpretable cross-attention network (ICAN), which is composed of nn.Embedding of FCS label-encoding vectors of SMILES of drugs and AA sequences of target proteins, cross-attention mechanisms, and a CNN output layer. ICAN decoded drug-related protein context features without any protein-related drug context features, achieving high prediction performance despite the plain attention mechanism. In comparison with seven state-of-the-art methods, ICAN provided the highest PRAUC for the imbalanced datasets (DAVIS and BindingDB). Statistical analysis of attention-weight matrixes revealed that some weighted attention sites corresponded to experimental binding sites, demonstrating the high interpretability achievable with ICAN.

bioinformatics↗

bioRxiv2022

Cross-attention PHV: Prediction of human and virus protein-protein interactions using cross-attention-based neural networks

Viral infections represent a major health concern worldwide. The alarming rate at which SARS-CoV-2 spreads, for example, led to a worldwide pandemic. Viruses incorporate genetic material into the host genome to hijack host cell functions such as the cell cycle and apoptosis. In these viral processes, protein-protein interactions (PPIs) play critical roles. Therefore, the identification of PPIs between humans and viruses is crucial for understanding the infection mechanism and host immune responses to viral infections and for discovering effective drugs. Experimental methods such as yeast two-hybrid assays and mass spectrometry are widely used to identify human-virus PPIs, but these experimental methods are time-consuming, expensive, and laborious. To overcome this problem, we developed a novel computational predictor, named cross-attention PHV, by implementing two key technologies of the cross-attention mechanism and a one- dimensional convolutional neural network (1D-CNN). The cross-attention mechanisms were very effective in enhancing prediction and generalization abilities. Application of 1D-CNN to the word2vec-generated feature matrices reduced computational costs, thus extending the allowable length of protein sequences to 9000 amino acid residues. Cross- attention PHV outperformed existing state-of-the-art models using a benchmark dataset and accurately predicted PPIs for unknown viruses. Cross-attention PHV also predicted human-SARS-CoV-2 PPIs with area under the curve values >0.95.

bioinformatics↗

bioRxiv2021

Stochastic comparison of synchronization in activator- and repressor-based coupled gene oscillators

Inside living cells, proteins or mRNA can show oscillations even without a periodic driving force. Such genetic oscillations are precise timekeepers for cell-cycle regulations, pattern formation during embryonic development in higher animals, and daily cycle maintenance in most organisms. The synchronization between oscillations in adjacent cells happens via intercellular coupling, which is essential for periodic segmentation formation in vertebrates and other biological processes. While molecular mechanisms of generating sustained oscillations are quite well understood, how do simple intercellular coupling produces robust synchronizations are still poorly understood? To address this question, we investigate two models of coupled gene oscillators - activator-based coupled oscillators (ACO) and repressor-based coupled oscillators (RCO) models. In our study, a single autonomous oscillator (that operates in a single cell) is based on a negative feedback, which is delayed by intracellular dynamics of an intermediate species. For the ACO model (RCO), the repressor protein of one cell activates (represses) the production of another protein in the neighbouring cell after a intercellular time delay. We investigate the coupled models in the presence of intrinsic noise due to the inherent stochasticity of the biochemical reactions. We analyze the collective oscillations from stochastic trajectories in the presence and absence of explicit coupling delay and make careful comparison between two models. Our results show no clear synchronizations in the ACO model when the coupling time delay is zero. However, a non-zero coupling delay can lead to anti-phase synchronizations in ACO. Interestingly, the RCO model shows robust in-phase synchronizations in the presence and absence of the coupling time delay. Our results suggest that the naturally occurring intercellular couplings might be based on repression rather than activation where in-phase synchronization is crucial.

systems biology↗

bioRxiv2021

LSTM-PHV: Prediction of human-virus protein-protein interactions by LSTM with word2vec

Viral infection involves a large number of protein-protein interactions (PPIs) between human and virus. The PPIs range from the initial binding of viral coat proteins to host membrane receptors to the hijacking of host transcription machinery. However, few interspecies PPIs have been identified, because experimental methods including mass spectrometry are time-consuming and expensive, and molecular dynamic simulation is limited only to the proteins whose 3D structures are solved. Sequence-based machine learning methods are expected to overcome these problems. We have first developed the LSTM model with word2vec to predict PPIs between human and virus, named LSTM-PHV, by using amino acid sequences alone. The LSTM-PHV effectively learnt the training data with a highly imbalanced ratio of positive to negative samples and achieved an AUC of 0.976 with an accuracy of 98.4% using 5-fold cross-validation. By using independent test dataset, we compared the LSTM-PHV with existing state-of-the-art PPI predictors including DeepViral. In predicting PPIs between human and unknown or new virus, the LSTM-PHV presented higher performance than the existing predictors when they were trained by multiple host protein-including datasets. LSTM-PHV learnt multiple host protein sequence contexts more efficiently than the DeepViral. Interestingly, learning of only sequence contexts as words presented remarkably high performances. Use of uniform manifold approximation and projection demonstrated that the LSTM-PHV clearly distinguished the positive PPI samples from the negative ones. We presented the LSTM-PHV online web server that is freely available at http://kurata35.bio.kyutech.ac.jp/.

bioinformatics↗

bioRxiv2021

RCGAToolbox: A real-coded genetic algorithm software for parameter estimation of kinetic models

SummaryKinetic modeling is essential in understanding the dynamic behavior of biochemical networks, such as metabolic and signal transduction pathways. However, parameter estimation remains a major bottleneck in the development of kinetic models. We present RCGAToolbox, software for real-coded genetic algorithms (RCGAs), which accelerates the parameter estimation of kinetic models. RCGAToolbox provides two RCGAs: the unimodal normal distribution crossover with minimal generation gap (UNDX/MGG) and real-coded ensemble crossover star with just generation gap (REXstar/JGG), using the stochastic ranking method. The RCGAToolbox also provides user-friendly graphical user interfaces. Availability and implementationRCGAToolbox is available from https://github.com/kmaeda16/RCGAToolbox under GNU GPLv3, with application examples. The user guide is provided in the Supplementary Material. RCGAToolbox runs on MATLAB in Windows, Linux, and macOS. Contactkmaeda@bio.kyutech.ac.jp Supplementary informationSupplementary Material is available at Bioinformatics online.

systems biology↗

bioRxiv2020

Virtual metabolic human dynamic model for pathological analysis and therapy design for diabetes

A virtual metabolic human model is a valuable complement to experimental biology and clinical studies, because in vivo research involves serious ethical and technical problems. I have proposed a multi-organ and multi-scale kinetic model that formulates the reactions of enzymes and transporters with the regulation of enzyme activities and hormonal actions under prandial and rest conditions. The model consists of 202 ordinary differential equations for metabolites with 217 reaction rates and 1132 kinetic parameter constants. It is the most comprehensive, largest and highly predictive model of the whole-body metabolism. Use of the model revealed the mechanisms by which individual disorders, such as steatosis, {beta} cell dysfunction and insulin resistance, were combined to cause type 2 diabetes. The model predicted a glycerol kinase inhibitor to be an effective medicine for type 2 diabetes, which not only decreased hepatic triglyceride but also reduced plasma glucose. The model also enabled us to rationally design combination therapy.

systems biology↗

bioRxiv2018

Improvement of the memory function by negative autoregulation of a mutual repression network in a stochastic environment

BackgroundCellular memory is a ubiquitous function of biological systems. By generating a sustained response to a transient inductive stimulus, often due to bistability, memory is central to the robust control of many important biological processes. However, our understanding of the origins of cellular memory remains incomplete. Stochastic fluctuations that are inherent to most biological systems have been shown to hamper memory function. Yet, how stochasticity changes the behavior of genetic circuits is generally not clear from a deterministic analysis of the network alone. Here, we apply deterministic rate equations, stochastic simulations, and theoretical analyses of Fokker-Planck equations to investigate how intrinsic noise affects the memory function in a mutual repression network.\n\nResultsWe find that the addition of negative autoregulation improves the persistence of memory in a small gene regulatory network by reducing stochastic fluctuations. Our theoretical analyses reveal that this improved memory function stems from an increased stability of the steady states of the system. Moreover, we show how the tuning of critical network parameters can further enhance memory.\n\nConclusionsOur work illuminates the power of stochastic and theoretical approaches to understanding biological circuits, and the importance of considering stochasticity to designing synthetic circuits with memory function.

systems biology↗