Here we demonstrate how we can use RGT-Viz for drawing a lineplots. This allows for example to inspect the ChIP-Seq signals around particular genomic regions, as PU.1. peaks. Before you proceed, please install RGT-Viz.
Download the example files
We will use the epigenetic data from dendritic cell development study  as example. There, we have ChIP-Seq data from the transcription factor PU.1. and histone modifications H3K4me3, H3K4me1 and H3K27me3 on four cellular states: multipotent progenitors (MPP), dendritic cell progenitors (CDP), common dendritic cells (cDC) and plamatocyte dendritic cells (pDC). For simplicity, here we only look at data from the differentiation transition CDP to cDC.
Please follow the following steps to download the necessary example files: genomic signals in bigwig (bw) format for histone modifications and PU.1, as well as peaks from PU.1 (bed files).
- Download the folder “example_RGT-Viz” from here.
- Execute the script “download_examples_RGT-viz.sh”
cd example_RGT-Viz sh download_examples_RGT-viz.sh
Now you have the files as described below:
example_RGT-Viz/ ├── data │ ├── cDC_H3K27me3.bw │ ├── cDC_H3K4me1.bw │ ├── cDC_H3K4me3.bw │ ├── cDC_PU1.bw │ ├── cDC_PU1_peaks.bed │ ├── CDP_H3K27me3.bw │ ├── CDP_H3K4me1.bw │ ├── CDP_H3K4me3.bw │ ├── CDP_PU1.bw │ └── CDP_PU1_peaks.bed ├── download_examples_RGT-viz.sh ├── Matrix_CDP_cDC.txt └── Matrix_CDP.txt
These files include the genomic signals of histone modifications (files with a .bw ending) and the genomic regions of PU.1 peaks (files with .bed endings).
Understand experimental matrix
Before we use the RGT-Viz, you must define an experimental matrix. This tab separated file includes information necessary for RGT to understand your data, i.e. file paths, protein measured in the ChIP-Seq experiment, type of file and so on.
For example “Matrix_CDP.txt” includes the files, which we need for finding the association of genomic signals on the genomic peaks of PU.1 transcription factor.
The first column (name) is a unique name for labeling the data; the second column indicate the type of experiment. Here we have either “regions” (genomic regions in bed format) or “reads” (genomic signals in bigwig or bam format). The third column is the file path to the data. You can include additional columns to annotate your data. In our example, the 4th column (factor) indicates the protein measured by the ChIP-Seq and the 5th collumn indicates the cell, where experiments were performed. You can add any more columns and the column names identify the feature.
After defining the experiment matrix, now you can simply run RGT-Viz under “example_RGT-Viz” directory by:
rgt-viz lineplot Matrix_CDP.txt -o results -t lineplot_CDP
- Matrix_CDP.txt is the experimental matrix which contains the design of the data;
- -o indicates the output directory;
- -t defines the title of this experiment.
This command will generate a directory “results” with figures and html pages.
You can check the result by opening results/index.html
Add one more cell type
Lineplot is designed to compare more categories of data. Here we show another example to include one more cell type, cDC.
rgt-viz lineplot Matrix_CDP_cDC.txt -o results -t lineplot_CDP_cDC -col cell -row regions -srow
- Matrix_CDP_cDC.txt is the experimental matrix which contains the design of the data;
- -col defines the way to group data in columns, here we use “cell”, which is one of the headers in Matrix_CDP_cDC.txt;
- -row defines the way to group data in rows, here we use “regions”;
- -sx shares the y-axis for the plots in the same row.
For better comparison of each genomic signal, we can also plot them in different way, such as:
rgt-viz lineplot Matrix_CDP_cDC.txt -o results -t lineplot_CDP_cDC_2 -c cell -row reads -col regions -sx
- -c defines the way to color the lines, here we use “cell” as the tag to show different cells in different colors;
- -row defines the way to group data in rows, here we use “reads”;
- -col defines the way to group data in columns, here we use “regions”.
Therefore, by changing the experimental matrix or the way to present, you can generate more complicated lineplot for comparison of your data across cell types, treatments, histone modification, or any other designs. RGT-Viz allows several other plots variants.
 Lin Q*, Chauvistre H*, Costa IG*, Mitzka S, Gusmao EG, Haenzelmann S, Baying B, Hennuy B, Smeets H, Hoffmann K, Benes V, Sere K, Zenke M, Epigenetic and Transcriptional Architecture of Dendritic Cell Development, Nucleic Acids Research, 43:9680-9693, [paper][data][genome tracks].