A hierarchical regulatory landscape during the multiple stages of EMT

This website is built using Shiny [1]. The (micro)RNA-seq analysis is performed using DESeq2 [2]. The microscopy images are displayed using EBImage [3].

Principal component analysis of (micro)RNA sequencing data

We use limma to estimate and remove batch effects from the regularised log (rlog) transformed gene expression profiles.

Time-course trajectories

We compare the ratio of two likelihoods; one for a model with both time and batch as explanatory variables versus one for a model with only batch information, in order to find genes whose differential expression across all samples is significantly better explained by time and batch than by batch alone.

Differential expression analysis

The p-values correspond to the likelihood ratio test explained above. However, the fold change at a reference time-point compared to a baseline timepoint can be extracted using the following drop-down menus.
The differential expression table refers to contrasting the perturbation with its control experiment, both sampled in biological duplicates after four days of TGF-beta.
The differential expression table refers to contrasting the perturbation with its control experiment, both sampled in biological duplicates after four days of TGF-beta.

Volcano plot

The volcano plot shows the differential expression of the selected gene across perturbation experiments (compared to control). The x-axis indicates log2 fold change, and the y-axis indicates -log10 of the adjusted p-value (Benjamini-Hochberg FDR correction). The dotted line corresponds to an adjusted p-value of 5%.

Microscopy screen images

From left to right; channel 1: phalloidin, channel 2: paxillin, channel 3: DAPI, channel 4: fibronectin

[1]: Winston Chang, Joe Cheng, JJ Allaire, Yihui Xie and Jonathan McPherson (2018). shiny: Web Application Framework for R. R package version 1.2.0. https://CRAN.R-project.org/package=shiny

[2]: Love, M.I., Huber, W., Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 Genome Biology 15(12):550 (2014)

[3]: Gregoire Pau, Florian Fuchs, Oleg Sklyar, Michael Boutros, and Wolfgang Huber (2010): EBImage - an R package for image processing with applications to cellular phenotypes. Bioinformatics, 26(7), pp. 979-981, 10.1093/bioinformatics/btq046