README.md

In shohre73/sigtools: Visualizing Genomic Signals

sigtools

Sigtools is an R-based visualization package, designed to enable the users with limited programming experience to produce statistical plots of continuous genomic data. It consists of several statistical visualizations that provide insights regarding the behavior of a group of signals in large regions – such as a chromosome or the whole genome – as well as visualizing them around a specific point or short region.

Third-Party Requirement

bwtool

bedGraphToBigWig

Installation

You can install the released version of sigtools from CRAN with:

install.packages("sigtools")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("shohre73/sigtools")

Input

Genomic signals are genomic coordinates associated with coverage measurements, probabilities, or scores. multi-column bedGraph is the primary input format for the Sigtools visual tasks.

#>         chr    start      end H3K4me1 H3K4me3   H3K9me3  H3K27ac  H3K27me3
#> 53846 chr21 10769000 10769200       0  0.0000 0.7089270 0.000000 0.0000000
#> 53847 chr21 10769200 10769400       0  0.0000 0.0334369 0.147998 0.1679040
#> 53848 chr21 10769400 10769600       0  0.0000 0.5602450 0.655519 0.0387464
#> 53849 chr21 10769600 10769800       0  0.0000 0.0000000 0.000000 0.0000000
#> 53850 chr21 10769800 10770000       0  0.1601 0.5736020 0.000000 0.0970146
#>        H3K36me3
#> 53846 0.3371940
#> 53847 0.0000000
#> 53848 0.0159367
#> 53849 0.1509530
#> 53850 0.0762994

Users can convert several bedGraph files to a multi-column bedGraph file at the desired resolution by sing Sigtools toMCbedGraph.

Example

Here are some examples of Sigtools application:

library(sigtools)

Distribution

SigTools generates several distribution plots such as Kernel Density Distribution and Empirical Cumulative Distribution. These plots assist in estimating data’s primary characteristics; namely the existence of multiple local maxima, the overall range, outliers.

Options for distribution plots are nozero, enrichment and percentile.

distribution("./test_data/E003-assays_chr21_bin200_bwtool.mcBedGraph", 
             outdir = "distribution", 
             plots = c("ecdf"), 
             header = TRUE, 
             percentile = 99, 
             enrichment = FALSE, 
             nozero = FALSE, 
             x_label = "", 
             y_label = "")

distribution("./test_data/E003-assays_chr21_bin200_bwtool.mcBedGraph", 
             outdir = "distribution", 
             plots = c("curve"), 
             header = TRUE, 
             percentile = 99, 
             enrichment = FALSE, 
             nozero = TRUE, 
             x_label = "", 
             y_label = "")

Autocorrelation

The autocorrelation plot displays the dependency of consecutive bins for each signal by recurrently calculating the correlation of a signal and itself when shifted.

autocorrelation("test_data/E003-assays_chr21_bin200_bwtool.mcBedGraph", 
                outdir = "autocorrelation", 
                header = TRUE, 
                resolution = 200L, 
                lag = 50L)

Correlation

The correlation gives you a heatmap of Pearson correlation for 2 sets of signals. For better comprehension, the association of variables is encoded in both color and area.

correlation(path_mulColBedG_1 = "test_data/E003-assays_chr21_bin200_bwtool.mcBedGraph", 
            header_1 = TRUE, 
            path_mulColBedG_2 = "test_data/E003-assays_chr21_bin200_bwtool.mcBedGraph", 
            header_2 = TRUE, 
            outdir = "correlation", 
            sep = '\t', 
            img_width = 10, 
            img_height = 10, 
            fontSize = 20)

Aggregation

The aggregation plot analyzes the behavior of a signal upon every occurrence of a specific element. Here, the enriched values of H3K36me3 are depicted over gene bodies of the human 21st chromosome.

aggregation(path_mulColBedG = "./test_data/E003-assays_chr21_bin200_bwtool.mcBedGraph",
            path_reg = "./test_data/Ensembl_chr21.gene_info",
            chr = 21,
            mode = "stretches",
            header = TRUE,
            sep = '\t',
            outdir = "aggregation",
            resolution = 200L,
            neighborhood = 20L,
            enrichment = TRUE)

shohre73/sigtools documentation built on April 2, 2021, 4:52 a.m.