Medical Sciences Group／Core LaboratoriesKamatani Laboratory
(Laboratory of Complex Trait Genomics)
-
Professor KAMATANI Yoichiro
Our laboratory focuses on analyzing human biological big data, such as genomics and omics, based on statistical genetics and learning theory. We aim to approach the question of what complex traits are through data analysis. Traits are the properties inherent to organisms, characterizing individuals. Examples of traits include physical appearance, personality, blood test values, or diseases. Traits not explainable by a few genes are referred to as complex traits. By analyzing data, we strive to elucidate biology and medicine, contributing to the advancement of medical care in Japan and nurturing talent engaged in research that serves this purpose.
- Research
keywords - genetics, genome-wide association study, polygenic risk score, genomics and omics, AI and machine learning
Genome wide association study
Genome-wide association study (GWAS) is a foundational technology for elucidating the genetic architecture of complex traits. Globally, many traits have been elucidated so far, and we participate in integrated analyses through international collaborative research or identify complex traits that have yet to be analyzed and conduct their GWAS. Recent target traits include stroke or peptic ulcer disease.
Features related to traits in the human genome, the blueprint of our molecular foundation, serve as clues to unravel the biological mechanisms of traits. We aim to contribute to the understanding of the traits through mechanistic data inference.
GWAS for peptic ulcer disease. More than 20 genetic loci are associated with disease.
Analysis of metabolomics and proteomics
The big data that can be analyzed from human populations extends beyond the genome to various omics data. In our lab, we are advancing the analysis of metabolome and proteome data obtained from analyzing serum samples in the BioBank Japan project. This not only helps in more concretely elucidating how genomic diversity impacts traits as biology, but also aims to contribute to the creation of global knowledge through participation in international collaborative research. Additionally, we investigate the predictive ability of disease onset using omics data that changes throughout a lifetime, not just the genome determined at birth.
Genomic analysis for brain vascular imaging
In the international genome-wide meta-analyses on cerebrovascular disorders, in which our laboratory also participated, it has been suggested that the genetic diversity of brain aneurysms is associated with disease through the diversity of gene expression in vascular endothelium, and the genetic diversity of stroke is related to the diversity of gene expression in vascular smooth muscle. In response to them, we will conduct a genomic analysis of cerebrovascular imaging to clarify how the condition of cerebral vessels is related to disease.
Deep learning analysis for spatial transcriptome
Pathological examinations, which involve collecting and staining samples of diseased tissue to identify characteristics under a microscope, are used in general clinical practice for cancer screening and other tests. In recent years, a method called spatial transcriptomics has been developed, which allows for the comprehensive measurement of gene expression at specific locations on pathological images, which was not previously possible.
As a collaborative effort within our department, we have developed a method called DeepSpaCE. This method addresses the challenges of super-resolution and three-dimensional completion of spatial transcriptomics. With this technique, researchers who are not experts in histology or pathology can interpret tissue section images in detail through gene annotation, and its application is expected in various fields including cancer research.
DeepSpaCE can predict spatial transcriptomic expression levels by using HE image as an input
References/papers
- He Y, Koido M, Shi M, Kamatani Y et al. East Asian-specific and cross-ancestry genome-wide meta-analyses provide mechanistic insights into peptic ulcer disease. Nat Genet 2023
- Mishra A, Koido M, Shi M, He Y, Kamatani Y et al. Stroke genetics informs drug discovery and risk prediction across ancestries. Nature 2022; 611: 115-23.
- Bakker M, Koido M, Kamatani Y et al. Genome-wide association study of intracranial aneurysms identifies 17 risk loci and genetic overlap with clinical risk factors. Nat Genet 2020; 52: 1303-13.
