Kota Matsui

Association of genetic variants with hypertension in a longitudinal population-based genetic epidemiological study

We previously identified 9 genes and chromosomal region 3q28 as susceptibility loci for Japanese patients with myocardial infarction, ischemic stroke, or chronic kidney disease by genome-wide or candidate gene association studies. In the present study, we investigated the possible association of 13 single nucleotide polymorphisms (SNPs) at these 10 loci with the prevalence of hypertension or their association with blood pressure (BP) in community-dwelling individuals in Japan. The study subjects comprised 6,027 individuals (2,250 subjects with essential hypertension, 3,777 controls) who were recruited into the Inabe Health and Longevity Study, a longitudinal genetic epidemiological study on atherosclerotic, cardiovascular and metabolic diseases. The subjects were recruited from individuals who visited the Health Care Center of Inabe General Hospital for an annual health checkup, and they are followed up each year (mean follow-up period, 5 years). Longitudinal analysis with a generalized estimating equation and with adjustment for age, gender, body mass index and smoking status revealed that rs2116519 of family with sequence similarity 78, member B (FAM78B; P=0.0266), rs6929846 of butyrophilin, subfamily 2, member A1 (BTN2A1; P= 0.0013), rs146021107 of pancreatic and duodenal homeobox 1 (PDX1; P=0.0031) and rs1671021 of lethal giant larvae homolog 2 (Drosophila) (LLGL2; P=0.0372) were significantly (P<0.05) associated with the prevalence of hypertension. Longitudinal analysis with a generalized linear mixed-effect model and with adjustment for age, gender, body mass index and smoking status among individuals not taking anti-hypertensive medication revealed that rs6929846 of BTN2A1 was significantly associated with systolic (P=0.0017), diastolic (P=0.0008) and mean (P=0.0005) BP, and that rs2116519 of FAM78B, rs146021107 of PDX1 and rs1671021 of LLGL2 were significantly associated with diastolic (P=0.0495), systolic (P=0.0132), and both diastolic (P=0.0468) and mean (0.0471) BP, respectively. BTN2A1 may thus be a susceptibility gene for hypertension.

Association of genetic variants with dyslipidemia and chronic kidney disease in a longitudinal population-based genetic epidemiological study

We previously identified 9 genes and chromosomal region 3q28 as susceptibility loci for myocardial infarction, ischemic stroke, or chronic kidney disease (CKD) in Japanese individuals by genome-wide or candidate gene association studies. In the present study, we examined the association of 13 polymorphisms at these 10 loci with the prevalence of hypertriglyceridemia, hyper-low-density lipoprotein (LDL) cholesterolemia, hypo-high-density lipoprotein (HDL) cholesterolemia, or CKD in community-dwelling Japanese individuals. The study subjects comprised 6,027 individuals who were recruited to the Inabe Health and Longevity Study, a longitudinal genetic epidemiological study of atherosclerotic, cardiovascular and metabolic diseases. The subjects were recruited from individuals who visited the Health Care Center at Inabe General Hospital for an annual health checkup, and they were followed up each year (mean follow-up period, 5 years). Longitudinal analysis with a generalized estimating equation and with adjustment for covariates revealed that rs6929846 of butyrophilin, subfamily 2, member A1 gene (BTN2A1) was significantly associated with the prevalence of hypertriglyceridemia (P=0.0001), hyper-LDL cholesterolemia (P=0.0004), and CKD (P=0.0007); rs2569512 of interleukin enhancer binding factor 3 (ILF3) was associated with hyper-LDL cholesterolemia (P=0.0029); and rs2074379 (P=0.0019) and rs2074388 (P=0.0029) of alpha-kinase 1 (ALPK1) were associated with CKD. Longitudinal analysis with a generalized linear mixed-effect model and with adjustment for covariates among all individuals revealed that rs6929846 of BTN2A1 was significantly associated with the serum concentrations of triglycerides (P=0.0011), LDL cholesterol (P=3.3×10−5), and creatinine (P=0.0006), as well as with the estimated glomerular filtration rate (eGFR) (P=0.0004); rs2569512 of ILF3 was shown to be associated with the serum concentration of LDL cholesterol (P=0.0221); and rs2074379 (P=0.0302) and rs2074388 (P=0.0336) of ALPK1 were shown to be associated with the serum concentration of creatinine. Similar analysis among individuals not taking any anti-dyslipidemic medication revealed that rs6929846 of BTN2A1 was significantly associated with the serum concentrations of triglycerides (P=8.3×10−5) and LDL cholesterol (P=0.0004), and that rs2569512 of ILF3 was associated with the serum concentration of LDL cholesterol (P=0.0010). BTN2A1 may thus be a susceptibility gene for hypertriglyceridemia, hyper-LDL cholesterolemia and CKD in Japanese individuals.

Parallel distributed block coordinate descent methods based on pairwise comparison oracle

This paper provides a block coordinate descent algorithm to solve unconstrained optimization problems. Our algorithm uses only pairwise comparison of function values, which tells us only the order of function values over two points, and does not require computation of a function value itself or a gradient. Our algorithm iterates two steps: the direction estimate step and the search step. In the direction estimate step, a Newton-type search direction is estimated through a block coordinate descent-based computation method with the pairwise comparison. In the search step, a numerical solution is updated along the estimated direction. The computation in the direction estimate step can be easily parallelized, and thus, the algorithm works efficiently to find the minimizer of the objective function. Also, we theoretically derive an upper bound of the convergence rate for our algorithm and show that our algorithm achieves the optimal query complexity for specific cases. In numerical experiments, we show that our method efficiently finds the optimal solution compared to some existing methods based on the pairwise comparison.

Empirical Bayes Estimation of Semi-parametric Hierarchical Mixture Models for Unbiased Characterization of Polygenic Disease Architectures

Genome-wide association studies (GWAS) suggest that the genetic architecture of complex diseases consists of unexpectedly numerous variants with small effect sizes. However, the polygenic architectures of many diseases have not been well characterized due to lack of simple and fast methods for unbiased estimation of the underlying proportion of disease-associated variants and their effect-size distribution. Applying empirical Bayes estimation of semi-parametric hierarchical mixture models to GWAS summary statistics, we confirmed that schizophrenia was extremely polygenic [~40% of independent genome-wide SNPs are risk variants, most within odds ratio (OR = 1.03)], whereas rheumatoid arthritis was less polygenic (~4 to 8% risk variants, significant portion reaching OR = 1.05 to 1.1). For rheumatoid arthritis, stratified estimations revealed that expression quantitative loci in blood explained large genetic variance, and low- and high-frequency derived alleles were prone to be risk and protective, respectively, suggesting a predominance of deleterious-risk and advantageous-protective mutations. Despite genetic correlation, effect-size distributions for schizophrenia and bipolar disorder differed across allele frequency. These analyses distinguished disease polygenic architectures and provided clues for etiological differences in complex diseases.

Obesity-related changes in clinical parameters and conditions in a longitudinal population-based epidemiological study

OBJECTIVES The purpose of the present study was to examine the association of body mass index (BMI) or obesity with various clinical parameters and conditions in a longitudinal population-based epidemiological study in Japan. METHODS Study subjects comprised 6027 community-dwelling individuals who were recruited to the Inabe Health and Longevity Study, a longitudinal genetic epidemiological study of atherosclerotic, cardiovascular, and metabolic diseases. Obesity was defined as BMI ≥25kg/m2. RESULTS Longitudinal analysis with the generalised linear mixed-effect model after adjustment for age showed that for men, BMI was significantly (P<0.0008) related to systolic, diastolic, and mean blood pressure and serum concentrations of high-density lipoprotein (HDL)-cholesterol and low-density lipoprotein (LDL)-cholesterol. For women, BMI was also significantly related to serum concentrations of triglycerides, HDL-cholesterol, and LDL-cholesterol. Longitudinal analysis with the generalised estimating equation with adjustment for age showed that in men, BMI was significantly (P<0.0012) associated with the prevalence of hypertension, type 2 diabetes mellitus, hypertriglyceridemia, hypo-HDL-cholesterolemia, decreased estimated glomerular filtration rate, and hyperuricemia. In women, BMI was also significantly associated with the prevalence of hypertension, type 2 diabetes mellitus, hypertriglyceridemia, and hyperuricemia. CONCLUSION Obesity has detrimental effects on various clinical parameters and conditions, resulting in increased risk of hypertension, dyslipidemia, type 2 diabetes mellitus, hyperuricemia, and chronic kidney disease.

Hierarchical Multiobjective Fuzzy Random Linear Programming Problems

Abstract In this paper, we propose an interactive decision making method for hierarchical multiobjective fuzzy random linear pro- gramming problems (HMOFRLP), in which multiple decision makers in a hierarchical organization have their own multiple objective linear functions with fuzzy random variable coefficients. To adress HMOFRLP, it is assumed that each decision maker has fuzzy goals for permissible probability levels in a fractile optimization model. Through a fuzzy decision, two types of membership functions of the original objective functions and the corresponding permissible probability levels are integrated, and a Pareto optimal solution concept is defined. A satisfactory solution is obtained from among a Pareto optimal solution set through the interaction with the decision makers, in which the hierarchical decision structure is reflected through the decision powers.

Multi-subgroup gene screening using semi-parametric hierarchical mixture models and the optimal discovery procedure: Application to a randomized clinical trial in multiple myeloma

This article proposes an efficient approach to screening genes associated with a phenotypic variable of interest in genomic studies with subgroups. In order to capture and detect various association profiles across subgroups, we flexibly estimate the underlying effect size distribution across subgroups using a semi-parametric hierarchical mixture model for subgroup-specific summary statistics from independent subgroups. We then perform gene ranking and selection using an optimal discovery procedure based on the fitted model with control of false discovery rate. Efficiency of the proposed approach, compared with that based on standard regression models with covariates representing subgroups, is demonstrated through application to a randomized clinical trial with microarray gene expression data in multiple myeloma, and through a simulation experiment.

Accuracy of the first interpretation of early brain CT images for predicting the prognosis of post-cardiac arrest syndrome patients at the emergency department

BackgroundEarly brain CT is one of the most useful tools for estimating the prognosis in patients with post-cardiac arrest syndrome (PCAS) at the emergency department (ED). The aim of this study was to evaluate the prognosis-prediction accuracy of the emergency physicians’ interpretation of the findings on early brain CT in PCAS patients treated by targeted temperature management (TTM).MethodsThis was a double-center, retrospective, observational study. Eligible subjects were cardiac arrest patients admitted to the intensive care unit (ICU) for TTM between April 2011 and March 2017. We performed the McNemar test to compare the predictive accuracies of the interpretation by emergency physicians and radiologists and calculated the kappa statistic for determining the concordance rate between the interpretations by these two groups.ResultsOf the 122 eligible patients, 106 met the inclusion criteria for this study. The predictive accuracies (sensitivity, specificity) of the interpretations by the emergency physicians and radiologists were (0.34, 1.00) and (0.41, 0.93), respectively, with no significant difference in either the sensitivity or specificity as assessed by the McNemar test. The kappa statistic calculated to determine the concordance between the two interpretations was 0.66 (0.48–0.83), which showed a good conformity.ConclusionsThe emergency physicians’ interpretation of the early brain CT findings in PCAS patients treated by TTM was as reliable as that of radiologists, in terms of prediction of the prognosis.

Association of renal function with clinical parameters and conditions in a longitudinal population‑based epidemiological study

The aim of the present study was to examine the association of renal function with clinical parameters and conditions in the general population. Study subjects comprised 6,027 community-dwelling individuals who were recruited to the Inabe Health and Longevity Study: A longitudinal genetic epidemiological study of atherosclerotic, cardiovascular and metabolic diseases. The cutoff value, which was used to divide the subjects into those with normal and those with low estimated glomerular filtration rate (eGFR), was 60 ml/min/1.73 m2. Bonferronis correction was applied to establish the statistical significance of the association. Longitudinal analysis using the generalized linear mixed-effect model, following adjustments for age and gender, revealed that the eGFR was significantly associated (P<0.0017) with serum levels of triglycerides, low-density lipoprotein cholesterol, uric acid, blood glycosylated hemoglobin content, fasting plasma glucose and body mass index. These parameters decreased curvilinearly with increases in eGFR. Furthermore, eGFR correlated positively with serum levels of high-density lipoprotein (HDL) cholesterol. Longitudinal analysis using the generalized estimating equation following adjustment for age and gender indicated a significant association (P<0.0024) between eGFR and prevalence of hypertension, type 2 diabetes mellitus, hypo-HDL cholesterolemia, hyperuricemia and obesity. Thus, low eGFR results in detrimental effects on various clinical parameters and conditions, resulting in increased risk of hypertension, dyslipidemia, type 2 diabetes mellitus, hyperuricemia and obesity.

Multiobjective Fuzzy Random Linear Programming Problems Based on E-Model and V-Model

In this paper, an interactive decision making method for multiobjective fuzzy random linear programming problems based on an expectation model (E-model) and a variance minimization model (V-model) is proposed. In the proposed method, it is assumed that the decision maker intends to not only maximize the expected degrees of possibilities that the original objective functions attain the corresponding fuzzy goals, but also minimize the standard deviations for such possibilities, and such fuzzy goals are quantified by eliciting the corresponding membership functions. Using the fuzzy decision, both the expected degrees of possibilities and the membership functions of the standard deviations are integrated, and an EV-Pareto optimality concept is introduced. In the integrated membership space, a satisfactory solution is obtained from among an EV-Pareto optimal solution set through the interaction with the decision maker.