A5055963311- Genetic Associations and Epidemiology
- Pancreatic function and diabetes
- Genetic Mapping and Diversity in Plants and Animals
- Genetic and phenotypic traits in livestock
- Statistical Methods and Inference
- Gene expression and cancer classification
- Diabetes and associated disorders
- Cannabis and Cannabinoid Research
- Advanced Statistical Methods and Models
- Autism Spectrum Disorder Research
- Genomic variations and chromosomal abnormalities
- Radiomics and Machine Learning in Medical Imaging
- Bioinformatics and Genomic Networks
- Bayesian Methods and Mixture Models
- Health, Environment, Cognitive Aging
- Skin Protection and Aging
- Genetics and Neurodevelopmental Disorders
- AI in cancer detection
- EEG and Brain-Computer Interfaces
- Xenotransplantation and immune response
- Biochemical Analysis and Sensing Techniques
- Data-Driven Disease Surveillance
- Hair Growth and Disorders
- Functional Brain Connectivity Studies
- Organ Transplantation Techniques and Outcomes
RIKEN Center for Advanced Intelligence Project
2018-2024
Nagasaki University
2021-2024
Tohoku University
2014-2021
Tohoku Medical Megabank Organization
2014-2021
Kurume University
2015-2018
Yamagata University
2009-2014
Newcastle University
2012
The Institute of Statistical Mathematics
2008-2011
Okayama University
2008
National Institute of Advanced Industrial Science and Technology
2003
The first step towards realizing personalized healthcare is to catalog the genetic variations in a population. Since dissemination of individual-level genomic information strictly controlled, it will be useful construct population-level allele frequency panels with easy-to-use interfaces. In Tohoku Medical Megabank Project, we sequenced nearly 4000 individuals from Japanese population and constructed an panel 3552 after removing related samples. called 3.5KJPNv2. It was by using standard...
Abstract Deep learning algorithms have been successfully used in medical image classification. In the next stage, technology of acquiring explainable knowledge from images is highly desired. Here we show that deep algorithm enables automated acquisition features diagnostic annotation-free histopathology images. We compare prediction accuracy prostate cancer recurrence using our algorithm-generated with diagnosis by expert pathologists established criteria on 13,188 whole-mount pathology...
In the Brassicaceae, intraspecific non-self pollen (compatible pollen) can germinate and grow into stigmatic papilla cells, while self-pollen or interspecific is rejected at this stage. However, mechanisms underlying selective acceptance of compatible remain unclear. Here, using a cell-impermeant calcium indicator, we showed that coat contains signaling molecules stimulate Ca(2+) export from cells. Transcriptome analyses stigmas suggested autoinhibited Ca(2+)-ATPase13 (ACA13) was induced...
Abstract Autism spectrum disorder (ASD) has phenotypically and genetically heterogeneous characteristics. A simulation study demonstrated that attempts to categorize patients with a complex disease into more homogeneous subgroups could have power elucidate hidden heritability. We conducted cluster analyses using the k-means algorithm number of 15 based on phenotypic variables from Simons Simplex Collection (SSC). As preliminary study, we conventional genome-wide association (GWAS) data set...
This paper develops smooth-threshold estimating equations that can automatically eliminate irrelevant parameters by setting them as zero. The resulting estimator enjoys the oracle property in sense of Fan & Li (2001), even estimators for which covariance assumption Wang Leng (2007) is violated, such Buckley–James estimator. Furthermore, be obtained without solving a convex optimization problem. A bic-type criterion tuning parameter selection also proposed. It shown achieves consistent model...
Abstract To solve major limitations in algorithms for the metabolite-based prediction of psychiatric phenotypes, a novel model depressive symptoms based on nonlinear feature selection machine learning, Hilbert–Schmidt independence criterion least absolute shrinkage and operator (HSIC Lasso) algorithm, was developed applied to metabolomic dataset with largest sample size date. In total, 897 population-based subjects were recruited from communities affected by Great East Japan Earthquake; 306...
Genome-wide gene-gene interaction analysis using single nucleotide polymorphisms (SNPs) is an attractive way for identification of genetic components that confers susceptibility human complex diseases. Individual hypothesis testing SNP-SNP pairs as in common genome-wide association study (GWAS) however involves difficulty setting overall p-value due to complicated correlation structure, namely, the multiple problem causes unacceptable false negative results. A large number than sample size,...
Deep learning algorithms have achieved great success in cancer image classification. However, it is imperative to understand the differences between deep and human approaches. Using an explainable model, we aimed compare learning-focused regions of magnetic resonance (MR) images with cancerous locations identified by radiologists pathologists. First, 307 prostate MR were classified using a well-established neural network without locational information cancers. Subsequently, assessed whether...
This study was designed to detect apoptosis in the human amnion and elucidate signalling pathway involved its regulation. Samples of were obtained from 34 women (weeks 11-42 gestation) studied using terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labelling (TUNEL) method with light microscopy (LM) transmission electron (TEM). Apoptotic regulators samples by immunohistochemistry caspase activity assay. The TUNEL LM demonstrated that percentage TUNEL-positive cells amniotic...
We investigated whether machine learning methods could potentially identify a subgroup of persons with autism spectrum disorder (ASD) who show vitamin B6 responsiveness by selected phenotype variables. analyzed the existing data from our intervention study 17 persons. First, we focused on signs and biomarkers that have been identified as candidates for indicators. Second, conducted hypothesis testing among these variables their combinations. Finally, further results conducting cluster...
Abstract The accuracy of previous genetic studies in predicting polygenic psychiatric phenotypes has been limited mainly due to the power distinguishing truly susceptible variants from null and resulting overfitting. A novel prediction algorithm, Smooth-Threshold Multivariate Genetic Prediction (STMGP), was applied improve genome-based by decreasing overfitting through selecting building a penalized regression model. models were trained using cohort 3685 subjects Miyagi prefecture validated...
Abstract Genetic maps are fundamental resources for linkage and association studies. A fine‐scale genetic map can be constructed by inferring historical recombination events from the genome‐wide structure of disequilibrium—a non‐random alleles among loci—by using population‐scale sequencing data. We a identified hotspots 10 092 551 bi‐allelic high‐quality autosomal markers segregating 150 unrelated Japanese individuals whose genotypes were determined high‐coverage (30×) whole‐genome...
The fused lasso penalizes a loss function by the L1 norm for both regression coefficients and their successive differences to encourage sparsity of both. In this paper, we propose Bayesian generalized modeling based on normal-exponential-gamma (NEG) prior distribution. NEG is assumed into difference coefficients. proposed method enables us construct more versatile sparse model than ordinary using flexible regularization term. Simulation studies real data analyses show that has superior...
Functional connectivity (FC) and neural excitability may interact to affect symptoms of autism spectrum disorder (ASD). We tested this hypothesis with network simulations, applied it functional magnetic resonance imaging (fMRI). A hierarchical recurrent embodying predictive processing theory was subjected a facial emotion recognition task. Neural simulations examined the effects FC on changes in representations by developmental learning, eventually ASD-like performance. Next, mapping each...
We develop a new genetic prediction method, smooth-threshold multivariate prediction, using single nucleotide polymorphisms (SNPs) data in genome-wide association studies (GWASs). Our method consists of two stages. At the first stage, unlike usual discontinuous SNP screening as used gene score our continuously screens SNPs based on output from standard univariate analysis for marginal each SNP. second predictive model is built by generalized ridge regression simultaneously screened with...
Age is known to influence the risk of both cerebral ischemic lesions and impaired cognitive function. Diabetes mellitus (DM) can also be associated with impairment. However, there has been no study neuropsychological performance in association glucose metabolism status same-aged, community-dwelling elderly persons. The present was performed clarify which domains are metabolism/DM whether independent lesions.A total 172 residents Takahata, Japan, all whom were 78 years old, evaluated multiple...
Biobanks increasingly collect, process and store omics with more conventional epidemiologic information necessitating considerable effort in data cleaning. An efficient outlier detection method that reduces manual labour is highly desirable.We develop an unsupervised machine-learning for detection, namely kurPCA, uses principal component analysis combined kurtosis to ascertain the existence of outliers. In addition, we propose a novel regression adjustment approach improve by systematic...