Biology subjects

Harada, K.

Publications and source records attributed to Harada, K..

At least 19 recordsLinked to original sources

bioRxiv2026

Two closely related RNA helicases, UAP56 and URH49, couple mRNA splicing and export pathways with their distinct complexes

Transcription and mRNA processing are integrated into appropriate mRNA export pathway(s) enabling fine-tuned gene expression. The closely related RNA helicases, UAP56 and URH49, form distinct complexes known as the apo-TREX and the apo-AREX complexes, respectively, which remodel into the ATP-TREX complexes, thus forming multiple mRNA export pathways. However, the mechanisms by which helicases and their complexes shape these pathways remain unclear. Here, we showed that UAP56 and URH49 play general roles in mRNA splicing ahead of export. Further analyses revealed three patterns of splicing regulation and their association with mRNA export: Type I, in which both helicases redundantly or synergistically regulate splicing within the same transcripts, associated with cooperative mRNA export; Type II, in which either helicase preferentially regulates splicing in distinct transcripts; and Type III, in which the helicases exert opposing splicing effects within the same transcripts. Type II and III regulation was associated with selective mRNA export. Moreover, we demonstrated that the distinct complex formation influenced the splicing patterns regulated by the two helicases. Our findings reveal an additional layer of transcriptome regulation mediated by UAP56, URH49, and their complexes, extending beyond their canonical roles in mRNA export.

molecular biology↗

bioRxiv2026

Structural insights into fosfomycin efflux by a streptococcal ABC transporter

Gram-positive bacteria encode a broad array of ABC transporters that mediate substrate translocation across the cell membrane, with some contributing to their survival under environmental stresses such as antimicrobial exposure. While several of these transporters have been shown to exhibit multidrug efflux activity, the functional roles of many others remain unknown. Here, using an efflux pump screen in the opportunistic human pathogen Streptococcus pneumoniae, we identified a previously uncharacterized type IV ABC transporter (FoeAB) that confers resistance to the antibiotic fosfomycin. We show that purified FoeAB mediates fosfomycin transport in a liposome-reconstituted system and provide evidence that it functions as a multidrug efflux pump with substrate preferences distinct from those of known efflux pumps. Furthermore, we present cryogenic electron microscopy (cryo-EM) structures of FoeAB in inward- and outward-facing states, which reveal conformational changes associated with nucleotide binding and identify residues important for substrate transport. Collectively, these findings expand the known repertoire of antibiotic-exporting ABC transporters in Gram-positive bacteria and provide structural insight into its transport mechanism. SIGNIFICANCEEfflux pumps often play major roles in antibiotic resistance in clinically important pathogens, yet many of these transporters remain poorly defined. In this study, we identify and mechanistically characterize FoeAB, a previously unrecognized type IV ABC transporter in streptococcal species that mediates efflux of the antibiotic fosfomycin. By combining functional assays with structural studies of FoeAB in multiple conformational states, we propose a model for how this transporter recognizes its substrate and harnesses nucleotide-driven conformational changes to mediate substrate export. Our findings point to a previously underappreciated efflux capabilities in Gram-positive bacteria, highlighting the need for better characterization of membrane transporters to clarify their roles in bacterial physiology and drug resistance.

microbiology↗

bioRxiv2026

A bone fragment-based protocol for molecular analysis of osteocyte-associated transcripts in human bone specimens

Osteocytes play a central role in bone remodeling, mineral metabolism, and skeletal homeostasis, but direct molecular analysis of human osteocytes remains technically challenging because they are embedded within the mineralized bone matrix. Surgically obtained human bone specimens provide valuable material for studying human bone biology; however, surface-associated cells, marrow-derived cells, and adherent soft tissues can confound downstream transcript analysis. Here, we describe a bone fragment-based protocol for preparing surgically obtained human bone specimens for molecular analysis of osteocyte-associated transcripts. The protocol consists of mechanical trimming, mincing into small bone fragments, repeated washing, and five sequential rounds of collagenase digestion to reduce non-osteocytic cellular components associated with the bone surface and marrow spaces. The remaining mineralized bone fragments are then frozen in liquid nitrogen, cryogenically pulverized, and lysed in TRIzol reagent for total RNA extraction. Histological validation using residual maxillary bone specimens showed that sequential collagenase digestion markedly reduced adherent soft tissue and extra-matrix nuclei while preserving osteocyte lacunar occupancy. This protocol provides a practical workflow for bone fragment-based RNA analysis focused on osteocyte-associated transcripts in human bone specimens. Specifications table O_TBL View this table: org.highwire.dtl.DTLVardef@1cec618org.highwire.dtl.DTLVardef@2f746forg.highwire.dtl.DTLVardef@1854247org.highwire.dtl.DTLVardef@1c26c1aorg.highwire.dtl.DTLVardef@1473a88_HPS_FORMAT_FIGEXP M_TBL C_TBL

cell biology↗

bioRxiv2026

Co-targeting an AMPK--MAPK axis reprograms CAFs and suppresses PDAC

Pancreatic ductal adenocarcinoma (PDAC) is a fatal cancer characterized by limited therapeutic options and a highly treatment-resistant tumor microenvironment. Beyond tumor-intrinsic genetic alterations, growing evidence indicates that host-microbiome interactions influence cancer progression through microbial metabolites. However, how microbiome-derived metabolites influence oncogenic signaling in PDAC remains unclear. Here, integrated profiling revealed a consistent reduction of the microbial metabolite acetic acid in fecal samples from treatment-naive patients with PDAC and in a genetically defined Drosophila model recapitulating key PDAC driver alterations. Acetic acid activates AMP-activated protein kinase, and pharmacological activation of this pathway together with inhibition of mitogen-activated protein kinase signaling suppressed tumor growth in fly and mouse models. Combined pathway targeting restored AMPK activity and suppressed cancer-associated fibroblast activation. These findings identify a microbiome-associated metabolic vulnerability in PDAC and suggest that coordinated targeting of metabolic and oncogenic signaling may restrain tumor progression and improve therapeutic strategies.

cancer biology↗

bioRxiv2025

A Scalable Toolkit for Modeling 3D Surface-based Brain Geometry

3D surface-based computational mapping is more sensitive to localized brain alterations in neurological, developmental and psychiatric conditions than traditional gross volumetric analysis, providing fine-scale 3D maps of a wide range of surface-based features. Here we introduce a scalable toolkit for large-scale computational surface analysis, with efficient algorithms for multisite data integration, statistical harmonization, accelerated multivariate statistics, and visualization. We showcase the utility of the toolkit by mapping subcortical shape variations and factors that affect them across 21 international samples from the ENIGMA Bipolar Disorder Working Group (N=3,373).

neuroscience↗

bioRxiv2025

Unsupervised feature computation-based feature selection robustly extracted resting-state functional connectivity patterns related to mental disorders

Research on biomarkers for predicting psychiatric disorders from resting-state functional connectivity (FC) is advancing. While the focus has primarily been on the discriminative performance of biomarkers by machine learning, identification of abnormal FCs in psychiatric disorders has often been treated as a secondary goal. However, it is crucial to investigate the effect size and robustness of the selected FCs because they can be used as potential targets of neurofeedback training or transcranial magnetic stimulation therapy. Here, we incorporated approximately 5,000 runs of resting-state functional magnetic resonance imaging from six datasets, including individuals with three different psychiatric disorders (major depressive disorder [MDD], schizophrenia [SCZ], and autism spectrum disorder [ASD]). We demonstrated that an unsupervised feature-computation-based feature selection method can robustly extract FCs related to psychiatric disorders compared to other conventional supervised feature selection methods. We found that our proposed method robustly extracted FCs with larger effect sizes from the validation dataset compared to different types of feature selection methods based on supervised learning for MDD (Cohens d = 0.40 vs. 0.25), SCZ (0.37 vs. 0.28), and ASD (0.17 vs. 0.16). We found 78, 69, and 81 essential FCs for MDD, SCZ, and ASD, respectively, and these FCs were mainly thalamic and motor network FCs. The current study showed that the unsupervised feature-computation-based feature selection method robustly identified abnormal FCs in psychiatric disorders consistently across datasets. The discovery of such robust FCs will contribute to understanding neural mechanisms as abnormal brain signatures in psychiatric disorders. Furthermore, this finding can aid in developing precise therapeutic interventions, such as neurofeedback training or transcranial magnetic stimulation therapy.

neuroscience↗

bioRxiv2025

Label Noise in Pathological Segmentation is Overlooked, Leading to Potential Overestimation of Artificial Intelligence Models

Artificial intelligence (AI) has transformed medical imaging, driving advancements in radiology and endoscopy. Semantic segmentation, a pixel-level technique crucial for delineating pathological features, has become a cornerstone of digital pathology. Pathology segmentation AI models are often trained using annotations generated by pathologists. Despite the meticulous care typically exercised, these annotations frequently contain empirical label noise. However, the specific types of label noise in pathology data and their impact on AI model training remain inadequately explored. This study systematically investigated the effects of label noise on the performance of pathology segmentation models. Using publicly available datasets and a breast cancer semantic segmentation dataset, modules were developed to simulate four types of artificial label noise at varying intensity levels. These datasets were used to train deep learning models with encoder-decoder architectures, and their performance was evaluated using metrics such as the Dice coefficient, precision, recall, and intersection over union. The results indicated that models were highly susceptible to overfitting label noise, particularly boundary-dependent noise such as dilation and shrinkage. Discrepancies were identified between apparent performance scores obtained under real-world conditions and true performance scores derived using clean test data. This overestimation risk was most pronounced for datasets containing boundary-altering noise. Furthermore, random combinations of noise types and levels significantly impaired model generalization. This study underscores the critical importance of addressing label noise in pathology datasets. It is proposed that future efforts focus on developing standardized methods for quantifying and mitigating label noise, along with creating robust benchmarks using noise-inclusive datasets. Enhancing annotation quality and addressing label noise can improve the reliability and generalizability of AI in pathology, facilitating broader clinical adoption.

bioinformatics↗

bioRxiv2024

An integrated and scalable rodent cage system enabling continuous computer vision-based behavioral analysis and AI-enhanced digital biomarker development

1Home cage monitoring enables continuous observation of animals in familiar environments. It has large utility in preclinical testing, mechanistic studies, animal husbandry, and the general practice of the Replacement, Reduction, Refinement (3R) principles. Despite its acknowledged utility, home cage monitoring has not been broadly adopted. This is mainly due to the complexity of the tasks that must be solved to have a successful system that includes hardware and sensor development, data management, machine vision expertise, behavioral expertise, support, and user training. Here, we describe the Digital In Vivo System (DIV Sys), a modern end-to-end system for video-based rodent home cage monitoring. The DIV Sys consists of a cloud-based study design, monitoring, display, and visualization app (DIV App), local hardware for data acquisition cages (DAX), a machine learning model for tracking mice (mHydraNet) optimized for speed and accuracy, a study display and visualization app, and an advanced behavior quantification workbench (DIV Data). The platform seamlessly manages terabytes of video data in the cloud and is built around enterprise-level security and data standards. Collaborative tools enable teams across geographical locations to work together. As a demonstration of its utility, we used DIV Sys to analyze over a century of mouse videos across multiple geographic locations. We also characterized home cage behavior of 8 mouse strains and carried out customized video analysis. Together, we present a scalable home cage monitoring system for advanced behavior quantification for the rodent research community.

animal behavior and cognition↗

bioRxiv2024

Chromosome-scale genome assembly of acerola (Malpighia emarginata DC.)

Acerola (Malpighia emarginata DC.) is a tropical evergreen shrub that produces vitamin C-rich fruits. Increasing fruit nutrition is one of the main targets of acerola breeding programs. Genomic tools have been shown to accelerate plant breeding even in fruiting tree species, which generally have a long life cycle; however, the availability of genomic resources in acerola, so far, has been limited. In this study, as a first step toward developing an efficient breeding technology for acerola, we established a chromosome-scale genome assembly of acerola using high-fidelity long-read sequencing and genetic mapping. The resultant assembly comprises 10 chromosome-scale sequences that span a physical distance of 1,032.5 Mb and contain 35,892 predicted genes. Phylogenetic analysis of genome-wide SNPs in 60 acerola breeding materials revealed three distinct genetic groups. Overall, the genomic resource of acerola developed in this study, including its genome and gene sequences, genetic map, and phylogenetic relationship among breeding materials, will not only be useful for acerola breeding but will also facilitate genomic and genetic studies on acerola and related species.

genomics↗

bioRxiv2024

Immediate glucose signaling transmitted via the vagus nerve in gut-brain neural communication

Sucrose consumption is influenced by certain gut-brain signaling mechanisms. Among these, one pathway involves neuropod cells, which form synaptic connections with the vagus nerve, leading to the immediate activation of central dopaminergic pathways. This study explored the role of the frontal cortex in its process. We found that the vagus nerves immediate activation is mediated by the sodium-glucose cotransporter 1 (SGLT1) of neuropod cells after the intragastric glucose injection in mice. Also, we showed that the involvement of both astrocytes and neurons in the frontal cortex via D2 and D1 dopamine receptors, respectively, by in vivo Ca2+ imaging. Finally, we revealed that psychological stress, which induces a reduction in sucrose preference, significantly diminishes the activation levels of both the vagus nerve and the frontal cortex. These findings highlight the role of a comprehensive gut-brain network in modulating sucrose preference, involving neuropod cells, the vagus nerve, and the frontal cortex. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/586971v1_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@1f28baeorg.highwire.dtl.DTLVardef@b053a9org.highwire.dtl.DTLVardef@449431org.highwire.dtl.DTLVardef@5d3ad9_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗

bioRxiv2024

Comprehensive evaluation of pipelines for diagnostic biomarkers of major depressive disorder using multi-site resting-state fMRI datasets

The objective diagnostic and stratification biomarkers developed with resting-state functional magnetic resonance imaging (rs-fMRI) data are expected to contribute to more effective treatment for mental disorders. Unfortunately, there are currently no widely accepted biomarkers, partially due to the large variety of analysis pipelines for developing them. In this study we comprehensively evaluated analysis pipelines using a large-scale, multi-site fMRI dataset for major depressive disorder (MDD) (1162 participants from eight imaging sites). We explored the combinations of options in four subprocesses of analysis pipelines: six types of brain parcellation, four types of estimations of functional connectivity (FC), three types of site difference harmonization, and five types of machine learning methods. 360 different MDD diagnostic biomarkers were constructed using the SRPBS dataset acquired with unified protocols (713 participants from four imaging sites) as a discovery dataset and evaluated with datasets from other projects acquired with heterogeneous protocols (449 participants from four imaging sites) for independent validation. To identify the optimal options regardless of the discovery dataset, we repeated the same procedure after swapping the roles of the two datasets. We found pipelines that included Glassers parcellation, tangent-covariance, no harmonization, and non-sparse machine learning methods tended to result in high classification performance. The diagnosis results of the top 10 biomarkers showed high similarity, and weight similarity was also observed between eight of the biomarkers, except two that used both data-driven parcellation and FC computation. We applied the top 10 pipelines to the datasets of other mental disorders (autism spectral disorder: ASD and schizophrenia: SCZ) and eight of the ten biomarkers showed sufficient classification performances for both disorders, except two pipelines that included Pearson correlation, ComBat harmonization and random forest classifier combination. HighlightsO_LIWe evaluated the analysis pipelines of rsFC biomarker development. C_LIO_LIFour subprocesses in them were investigated with two multi-site datasets. C_LIO_LIGlassers parcellation, tangent covariance, and non-sparse methods were preferred. C_LIO_LIThe weight patterns of eight of the top 10 biomarkers showed high commonality. C_LIO_LIEight of the top 10 pipelines were successful for developing SCZ/ASD biomarkers. C_LI

neuroscience↗

bioRxiv2022

Exocytotic dynamics of glucagon-like peptide-1 from enteroendocrine L cell line is regulated by actin polymerization

Stimulus-secretion coupling of glucagon-like peptide-1 (GLP-1) from enteroendocrine L cells is important for glucose homeostasis. Although intracellular second messengers including Ca2+ and cAMP, and cellular structures including actin cytoskeleton play roles in induction of exocytosis of GLP-1 granules, little is known about the specific part in the process of exocytosis in which they are involved. Here we explored the role of those molecules by live-cell imaging with mouse L cell line GLUTag cells, and used two stimuli: deoxycholic acid (DCA) and high K+. DCA increased both intracellular Ca2+ and cAMP levels, while high K+ only increased Ca2+. We next monitored a single exocytosis of GLP-1 granules and found that, during the first 10 minutes of stimulation, both stimuli mainly induced the exocytosis from the predocked granules with the plasma membrane before stimulation or granules immediately fused to the plasma membrane without docking. Furthermore, inhibition of actin polymerization suppressed the proportion of exocytosis by the predocked granules. These results suggest that the exocytotic process of GLP-1 granules is determined by interaction with F-actin upon the increase of either Ca2+ or cAMP. Summary statementExocytotic process of glucagon-like peptide-1 granules from a mouse enteroendocrine L cell line is regulated by actin polymerization immediately after elevation of intracellular Ca2+ or cAMP levels.

cell biology↗

bioRxiv2020

Common brain networks between major depressive disorder and symptoms of depression that are validated for independent cohorts

Many studies have highlighted the difficulty inherent to the clinical application of fundamental neuroscience knowledge based on machine learning techniques. It is difficult to generalize machine learning brain markers to the data acquired from independent imaging sites, mainly due to large site differences in functional magnetic resonance imaging. We address the difficulty of finding a generalizable major depressive disorder (MDD) brain network markers which would distinguish patients from healthy controls (a classifier) or would predict symptom severity (a prediction model) based on resting state functional connectivity patterns. For the discovery dataset with 713 participants from 4 imaging sites, we removed site differences using our recently developed harmonization method and developed a machine learning MDD brain network markers. The classifier achieved 70% generalization accuracy, and the prediction model moderately well predicted symptom severity for an independent validation dataset with 449 participants from 4 different imaging sites. Finally, we found common 2 functional connections between those related to MDD diagnosis and those related to depression symptoms. The successful generalization to the perfectly independent dataset acquired from multiple imaging sites is novel and ensures scientific reproducibility and clinical applicability.

neuroscience↗

bioRxiv2019

DLDG1, the Arabidopsis Homolog of the Cyanobacterial H+-Extrusion Protein, Regulates Non-Photochemical Quenching in Chloroplasts

Plants convert solar energy into chemical energy through photosynthesis, which supports almost all life activities on earth. Because the intensity and quality of sunlight can change dramatically throughout the day, various regulatory mechanisms help plants adjust their photosynthetic output accordingly, including the regulation of light energy accumulation to prevent the generation of damaging reactive oxygen species. Non-photochemical quenching (NPQ) is a regulatory mechanism that dissipates excess light energy, but how it is regulated is not fully elucidated. Herein, we report a new NPQ-regulatory protein named Day-Length-dependent Delayed-Greening1 (DLDG1). The Arabidopsis DLDG1 associates with the chloroplast envelope membrane, and the dldg1 mutant had a large NPQ value compared with wild type. The mutant also had a pale-green phenotype in developing leaves but only under continuous light; this phenotype was not observed when dldg1 was cultured in the dark for [≥]8 h per day. DLDG1 is a homolog of the plasma-membrane-localizing cyanobacterial proton-extrusion-protein A that is required for light-induced H+ extrusion, and also shows similarity in its amino-acid sequence to that of Ycf10 encoded in the plastid genome. Arabidopsis DLDG1 enhances the growth-retardation phenotype of the Escherichia coli K+/H+ antiporter mutant, and the everted membrane vesicles of the E. coli expressing DLDG1 show the K+/H+ antiport activity. Our findings suggest that DLDG1 functionally interacts with Ycf10 to control H+ homeostasis in chloroplasts, which is important for the light-acclimation response, by optimizing the extent of NPQ.

plant biology↗

bioRxiv2019

Cell-of-Origin Analysis of Metastatic Gastric Cancer Uncovers the Origin of Inherent Intratumor Heterogeneity and a Fundamental Prognostic Signature

Intra-tumoral heterogeneity (ITH) is the fundamental property of cancer, however, the origin of ITH remains poorly understood. Here we performed single-cell RNA sequencing of peritoneal carcinomatosis (PC) from 20 patients with advanced gastric adenocarcinoma (GAC), constructed a transcriptome map of 45,048 PC cells, determined the cell-of-origin of each tumor cell, and incisively explored ITH of PC tumor cells at single-cell resolution. The links between cell-of-origin and ITH was illustrated at transcriptomic, genotypic, molecular, and phenotypic levels. This study characterized the origins of PC tumor cells that populate and thrive in the peritoneal cavity, uncovered the diversity in tumor cell-of-origins and defined it as a key determinant of ITH. Furthermore, cell-of-origin-based analysis classified PC into two cellular subtypes that were prognostic independent of clinical variables, and a 12-gene prognostic signature was then derived and validated in multiple large-scale GAC cohorts. The prognostic signature appears fundamental to GAC carcinogenesis/progression and could be practical for patient stratification.

cancer biology↗

bioRxiv2019

Processing of a single ribonucleotide embedded into DNA by human nucleotide excision repair and DNA polymerase η

DNA polymerases often incorporate non-canonical nucleotide, i.e., ribonucleoside triphosphates into the genomic DNA. Aberrant accumulation of ribonucleotides in the genome causes various cellular abnormalities. Here, we show the possible role of human nucleotide excision repair (NER) and DNA polymerase {eta} (Pol {eta}) in processing of a single ribonucleotide embedded into DNA. We found that the reconstituted NER system can excise the oxidized ribonucleotide on the plasmid DNA. Taken together with the evidence that Pol {eta} accurately bypasses a ribonucleotide, i.e., riboguanosine (rG) or its oxidized derivative (8-oxo-rG) in vitro, we further assessed the mutagenic potential of the embedded ribonucleotide in human cells lacking NER or Pol {eta}. A single rG on the supF reporter gene predominantly induced large deletion mutations. An embedded 8-oxo-rG caused base substitution mutations at the 3-neighboring base rather than large deletions in wild-type cells. The disruption of XPA, an essential factor for NER, or Pol {eta} leads to the increased mutant frequency of 8-oxo-rG. Furthermore, the frequency of 8-oxo-rG-mediated large deletions was increased by the loss of Pol {eta}, but not XPA. Collectively, our results suggest that base oxidation of the embedded ribonucleotide enables processing of the ribonucleotide via alternative DNA repair and damage tolerance pathways.

genetics↗

bioRxiv2019

Spatial structures of fungal DNA assemblages revealed with eDNA metabarcoding in a forest river network in western Japan

Growing evidence has revealed high diversity and spatial heterogeneity of fungal communities including in local habitats in terrestrial ecosystems. These findings highlight the considerable sampling effort, analysis time, and costs required for the investigation of fungal diversity over large spatial scales. Recently, the analysis of environmental DNA in river water has been undertaken to study the biodiversity of organisms, such as animals and plants, in both aquatic and terrestrial habitats. However, previous studies have not investigated the spatial structure of fungal DNA assemblages in river water. In the present study, we investigate fungal DNA assemblages and their spatial structure using environmental DNA metabarcoding in water across different branches of river over forest landscapes. The river water was found to contain both phylogenetically and functionally diverse fungal DNA, including aquatic and terrestrial fungi, such as plant decomposers and mycorrhizal fungi. These fungal DNA assemblages were more similar within, rather than between, branches. In addition, the assemblages were more similar between spatially closer branches. These results imply that information on the terrestrial and aquatic fungal compositions of watersheds, and therefore their spatial structure can be obtained by investigating the fungal DNA assemblages in river water.

ecology↗

bioRxiv2018

Customizing material into embodied cap by sponge crab

Animal personality is considered to be developed through complex interactions of an individual with its surrounding environment. How can we quantify the personality of an individual? Quantifying intra- and inter-individual variability of behavior, or individual behavioral type, appears to be a prerequisite in the study of animal personality. We propose a statistical method from a predictive point of view to measure the appropriateness of our assumption of individual behavior in repeatedly measured behavioral data from several individuals. For a model case, we studied the sponge crab Lauridromia dehaani known to make and carry a cap from a natural sponge for camouflage. Because a cap is most likely to be rebuilt and replaced repeatedly, we hypothesized that each individual crab would grow a unique behavioral type and it would be observed under an experimentally controlled environmental condition. To test the hypothesis, we conducted behavioral experiments and employed a new Bayesian model-based comparison method to examine whether crabs have individual behavioral types in the cap making behavior. Crabs were given behavioral choices by using artificial sponges of three different sizes. We modeled the choice of sponges, size of the trimmed part of a cap, size of the cavity of a cap, and the latency to produce a cap, as random variables in 26 models, including hierarchical models specifying the behavioral types. In addition, we calculated the marginal-level widely applicable information criterion (mWAIC) values for hierarchical models to evaluate and compared them with the non-hierarchical models from the predictive point of view. As a result, the crabs of less than about 9 cm in size were found to make caps from the sponges. The body size explained the behavioral variables namely, choice, trimmed cap characteristics, and cavity size, but not latency. Furthermore, we captured the behavioral type as a probabilistic distribution structure of the behavioral data by comparing WAIC. Our statistical approach is not limited to behavioral data but is also applicable to physiological or morphological data when examining whether some group structure exists behind fluctuating empirical data.

animal behavior and cognition↗