December 9, 2019 - Bioconductor conference, Brussels.


  • Goal: customized and semi-automated analysis of microarray and bulk RNAseq data

  • Authors: Laure Cougnaud, Kirsten Van Hoorde, Michela Pasetto, Katarzyna Górczak, Yingjie Zhang

  • Tool: R package transcriptomicsTemplates and associated templates in RMarkdown format

How it works

The user copies the original template and modifies towards a personal RMarkdown master document containing analysis-specific input parameters.

Running the RMarkdown master document:

  • calls other child templates contained in the package (do not need manual modifications)

  • creates:

    • an html report file associated to the Rmd template
    • objects/graphs folders with results/visualizations

Available analyses

  • unsupervised analysis (required)
  • differential expression analysis (optional)
  • overlap analysis (optional)
  • pairwise comparison analysis (optional)

Example dataset

  • ALL dataset
  • microarrays from patients with acute lymphoblastic leukemia
  • covariates of interest:
    • gender: male/female
    • different disease type: B-cell/T-cell
    • different disease stage: 1 to 4
    • remission achieved or refractory
  • data from: Gene expression profile of adult T-cell acute lymphocytic leukemia identifies distinct subsets of patients with different response to therapy and survival, Chiaretti, et al. ,Blood type, 2004, 103

Unsupervised analysis

For a first glance at the data: spectral map of the entire dataset or a subset of samples of interest:

  • with covariates-related aesthetics
  • with different levels of annotation: sample, gene, gene set
  • visualize genes/show table of pathways with highest score

Spectral maps with gene and pathway annotation

Differential expression analysis

  • For microarray: limma

  • For RNAseq: with edgeR and DESeq2

  • several analyses of interest

  • need to specify input model (design matrix, lmFit and eBayes)

  • for each analysis, results are reported with:

    • volcano plots
    • top affected genes
    • top affected pathways

Top affected genes