Using heatmaply with non-centred RNAseq heatmaps (PAM50 genes)

Author: Alan O’Callaghan (alan.b.ocallaghan@gmail.com)

1 Introduction

This vignette illustrates the clustering of non-centered breast cancer RNAseq data, similar to the centered data. shown in the main biological data vignette in this package.

pam50_genes <- intersect(pam50_genes, rownames(raw_expression))
raw_pam50_expression <- raw_expression[pam50_genes, ]
voomed_pam50_expression <- voomed_expression[pam50_genes, ]

log_raw_mat <- log2(raw_pam50_expression + 0.5)

heatmaply(t(log_raw_mat), 
    row_side_colors = tcga_brca_clinical,
    showticklabels = c(TRUE, FALSE),
    fontsize_col = 7.5,
    col = gplots::bluered(50),
    main = 'Pre-normalisation log2 counts, PAM50 genes',
    plot_method = 'plotly')

heatmaply_cor(cor(log_raw_mat), 
    row_side_colors = tcga_brca_clinical,
    showticklabels = c(FALSE, FALSE),
    main = 'Sample-sample correlation based on log2-transformed PAM50 gene expression',
    plot_method = 'plotly')