Skip to main content
Fig. 1 | BMC Medical Genomics

Fig. 1

From: Novel feature selection method via kernel tensor decomposition for improved multi-omics data analysis

Fig. 1

Schematic representation of HBV vaccination data analysis. Analysis starts from the center, moves to the right, comes back to the center, and then moves to the left. The cyan rectangle annotated as “methylation” is \(x_{i_1j_1j_2}\), the yellow rectangle annotated as “gene” is \(x_{i_2j_1j_2}\), the green rectangle annotated as “WBC” is \(x_{i_3j_1j_2}\), and the magenta rectangle annotated as “Plasma” is \(x_{i_4j_1j_2}\). The four tilted cubes to the right of these four rectangles are \(x_{{k}j_1j_2j'_1j'_2}\), whose correspondence with \(x_{i_kj_1j_2}\) is indicated by the same color. The tilted cubes colored by layers to the right of the four tilted cubes represent the bundle of \(x_{{k}j_1j_2j'_1j'_2}\). The right-most figure with a blue cube annotated as “G” at the center corresponds to TD shown in Eq. (16). The four colored rectangles to the left of the four colored and annotated rectangles represent the singular-value vectors computed by Eq. (17). Genes are selected from these singular-value vectors using P values computed by Eq. (18). For methylation, transcription factors (TFs) are further selected by Enricher using the selected genes (Table 3). The selected genes and TFs are then uploaded to Enrichr to validate the biological reliability (the left-most figure with color gradation)

Back to article page