1D-O1-1545 PDF
メタボロームデータ解析のためのケモメトリックス手法の開発
In metabolome data analysis, multivariate analysis such as principal component analysis (PCA) and partial least squares (PLS) have been widely applied. However, these methods were not able to select metabolites by using statistical criteria. We developed novel methods to select statistical significant metabolites by using statistical hypothesis testing of factor loading in PCA and PLS. And, we proposed PLS-rank order of groups (PLS-ROG) that was able to reflect rank order of groups in PLS subspace.