Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Feature-specific quantile normalization and feature-specific mean–variance normalization deliver robust bi-directional classification and feature selection performance between microarray and RNAseq data

Fig. 1

Study overview. Feature Specific Quantile Normalization and Feature Specific Mean Variance Normalization are two methods of cross-platform normalization that allow integration and/or comparison of microarray and RNAseq data. These methods use either quantile normalization or mean and variance matching at a gene-specific level to match gene expression data to a target distribution. These methods are bidirectional, in that microarray data can match the RNAseq distribution and vice versa. Here, we evaluate the validity of using FSQN and/or FSMVN in the context of supervised machine learning classifiers for molecular subtyping using cross-platform data. We compare the model accuracy to the unnormalized log2 gene expression data of the target/training distribution. We also evaluate whether FSQN and/or FSMVN is a valid method in the context of feature selection

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us