Current projects (2013-)

Variance component analysis using a linear mixed model with student t distribution allowing us to discern the relative importance of alternative sources of variation in multiple RNA-seq samples
Previous Projects
A Bayesian gaussian mixture model for genotyping copy number polymorphisms using SNP array signal intensity data.
An extention of the textile plot for visualizing haplotype structure of multi-allelic loci.
Principal component analysis (PCA) based normalization algorithm for signal intensities of the oligonucreotide assay.
Principal component analysis (PCA) based prediction model for population structure analysis with genome-wide SNPs.
An application of the textile plot in genetics and genomics fields. The textile plot could accentuate various linkage disequilibrium patterns as well as haplotype structures among SNP genotype markers. A new optimization algorithm has also been developed so as to handle a number of SNPs in genome-wide scale.
PhD and Masters Projects
Textile Plot, 2004-2007
A cutting edge graphical representation for multivariate data sets. It can visualize both numerical and categorical data or mixture of those in one display to seek for complex linear relationships among variables.
Necessary and sufficient descriptions for multivariate data sets that enable us to analyse data efficiently and effectively. We have defined the description rule in XML, proposed a concept of interdatabase and developed a server client system implemented in JAVA language.
DandDR, 2003
A client system of the DandD server client system that aims to import data from the DandD environment to the R environment.
Papers
Kumasaka N, Nakamura Y & Kamatani N (2010) The Textile Plot: a new linkage disequilibrium display of multiple-single nucleotide polymorphism genotype data. PLoS ONE 5(4): e10207. doi:10.1371/journal.pone.0010207.
Kumasaka N and Shibata R (2008) High-dimensional data visualisation: The textile plot. Computational Statistics & Data Analysis, 52(7):3616-44.