Contents

Package: Imetagene
Authors: Audrey Lemacon audrey.lemacon.1@ulaval.ca, Charles Joly Beauparlant charles.joly-beauparlant@crchul.ulaval.ca, Arnaud Droit arnaud.droit@crchuq.ulaval.ca
Modified: 9 october, 2015
Compiled: Mon Apr 24 20:23:27 2017
License: Artistic-2.0 | file LICENSE

1 Introduction

Imetagene is a graphical user interface for metagene that is based on the shiny package and is displayed in a web browser.
This interface is based on the last version of metagene (>= 2.1.31).

Metagene package “produces metagene-like plots to compare the behavior of DNA-interacting proteins at selected groups of features. A typical analysis can be done in vicinity of transcription start sites (TSS) of genes or at any regions of interest (such as enhancers). Multiple combinations of group of features and/or group of bam files can be compared in a single analysis. Bootstraping analysis is used to compare the groups and locate regions with statistically different enrichment profiles. In order to increase the sensitivity of the analysis, alignment data is used instead of peaks produced with peak callers (i.e.: MACS2 or PICS). The metagene package uses bootstrap to obtain a better estimation of the mean enrichment and the confidence interval for every group of samples.” - (go to metagene vignette)

Using this interface you can easily load your files (BAM et BED), build your experimental design, generate your coverage matrices and finally plot your metagene-like plots.

2 Organization overview

The interface is divided in four parts : Inputs, Design, Matrix and Plot.
A classical usage of the interface would be to first fill the Inputs part.

Once the input is filled, you can build your experimental design in the Design section. Thanks to the design option, multiple combinations of group of bam files can be compared in a single analysis.

You can then generate the coverages matrices according to the design if given in the Matrix panel.
Finally, the metagene plot are produced using the ggplot2 package in the Plot part.

2.1 Inputs

You can work with an existing metagene object previously generated by metagene (>= 2.1.31) or through Imetagene. The file must be a RData file (with extention .RData, .Rda, .RDATA or .RDA)

load existing metagene object

You also create a new metagene object. To do so, you have to provide the BAM files you want to use and the regions you want to study in BED format.

There is no hard limit in the number of BAM files that can be included in an analysis (but with too many BAM files, memory may become an issue). BAM files must be indexed. For instance, if you use a file names file.bam, a file named file.bam.bai must be present in the same directory.

To compare custom regions of interest, it is possible to use a list of one or more BED files. BED, narrowPeak and broadPeak format are supported (extention .bed, .BED, .narrowPeak and .broadPeak).

create a new metagene object

You can then save the metagene object for a future usage. Once the metagene is loaded or freshly created, you can go to the next step : Design creation.

2.2 Design

A design file is a tab-delimited file (.tsv) that describes one or more experiments. An experiment can contain one or more replicates and controls. The first column of a design file contains the list of bam names available in the current metagene object. The following columns correspond to the experiments. They must have a unique name and the possible values are 0 (ignore), 1 (chip) or 2 (control).

If you are using a loaded metagene object which already contains a design, it will be displayed in the Current design tab. If you load or create a new design it will override the existing one.

You can load an existing design file in the Load existing design file tab. If the file is well formated, it will be loaded and displayed the Current design tab.

load an existing design

To create a new design, you can go in the Create new design tab. You will have create one experiment at a time by :

create new experiment

Once it is done, you can add the experiment. The updated design will be displayed the Current design tab.

display design

You can then save the design in csv file for a further use.

2.3 Matrix

To produce the metagene plot, the coverages must be converted in a matrix where the columns represent the positions and the rows the regions. Furthermore, to reduce the computation time during the following steps, the positions are also binned.

If you are using a loaded metagene object which already contains processed matrices, they will be displayable in the Current matrix subset tab. If you produce new matrices, they will override the existing ones.

We can control the size of the bin either with the bin_size argument. By default, a bin_count of 100 will be used during this step.

We can also use the design we produced earlier to remove background signal and combine replicates.

display design

The Current matrix subset tab give you an overview of the generated matrices as heatmaps. Thanks to the package d3heatmap, you can hover the mouse pointer over a cell to show details, drag a rectangle to zoom, and click row/column ticks to highlight.

display design

You can then save the updated metagene object (which now contains a design and matrices) for a future usage.

2.4 Plot

The metagene plot requires a data.frame as input. The data.frame is computed in background and the values of the ribbon are calculated. By default, metagene uses “bootstrap” to obtain a better estimation of the mean of enrichment for every positions in each group of regions/BAM files.

The data.frame will be then used to plot the calculated values using ggplot2. We show a subset of the regions by using the Regions to display and Experiments to display parameter. The Regions to display correspond to the names of the regions used during the initialization. The Experiments to display will vary depending if a design was added. If no design was added, this param correspond to the BAM name or BAM filenames. Otherwise, we have to use the names of the columns from the design.

Here is the plot with the usage of a design

display design

And the same data without a design

display design