Large-scale cohort studies have been established in many countries. From these healthcare big data, there is a need to gain new insights such as identification of unknown risk factors and prediction of risk of developing disease, and to give back to society. Extensive data makes it difficult to perform comprehensive and detailed analysis by human power, but the data can be analyzed efficiently using advanced theories and methods of statistical science and machine learning. My main research interests are sparse modeling and other methods that are easily interpreted and also have high accuracy.

My research focuses on the development and practice of methodologies to investigate the relationship between human diseases and genes. In particular, I am studying ultrahigh dimensional genomic data such as genome-wide SNPs (single nucleotide polymorphisms) data. For example;

• Genome-wide association study
• Gene × gene interaction analysis
• Gene × environment interaction analysis
• Prediction of risk of developing disease

Using statistical science, biostatistics, and machine learning, I work on a variety of statistical genetics problems ranging from standard to advanced analysis of human genomic data.