marge

Overview

The aim of marge is to provide an R API to HOMER for the analysis of genomic data, utilizing a tidy framework to accelerate organization and visualization of analyses.

Installing HOMER

First, running marge requires having a working installation of HOMER on your computer. Please see the HOMER website for more information on installing and configuring HOMER and to learn more about the methodology. In particular, note that you should install your desired genomes in addition to installing HOMER using the ./configureHomer.pl script.

Note that working with a conda installation of HOMER is not well tested at this time. A potential workaround is below. I recommend installing directly from source.

Installing marge

To install the latest development version of marge, simply do:

devtools::install_github('robertamezquita/marge', ref = 'master')

Before Running marge

While marge will do its best to find HOMER, there are certain environments where it will not be able to do so, specifically, with regards to RStudio and conda installs of HOMER. In these cases, a custom path to HOMER can be provided if it is not found by the package's utilities by setting options('homer_path' = "/path/to/homer-4.10"). This can be set in your ~/.Rprofile so it loads the correct path automagically each time.

Usage

marge is currently semi-stable. The package currently includes the ability to:

• Run a motif analysis: find_motifs_genome() - runs the HOMER script findMotifsGenome.pl via R, and outputs a results directory in the default HOMER style
• Read in results: read_*_results() - read in either denovo or known enriched motifs with the read_denovo_results() or read_known_results() functions, pointing to the HOMER directory that was created in the previous step. The read_* functions produce tibbles summarizing the motif enrichment results into a tidy format for easier visualization and analysis. See the reference pages of each for more details.
• Write motifs in HOMER compatible format with write_homer_motif()
• Find specific motif instances across regions using supplied PWMs with find_motifs_instances() and read in the results with read_motifs_instances()
• Access the HOMER database of known motifs by inspecting the HOMER_motifs object

Further details can be found in the associated vignette, describing installation and typical workflows encompassing basic/advanced usage schemas.

Compared to HOMER alone

Like the actual Homer Simpson, HOMER is made better with the addition of marge. With the continually increasing throughput in conducting sequencing analysis, marge provides a native R framework to work from end to end with motif analyses - from processing to storing to visualizing these results, all using modern tidy conventions.

---

Contact

• Robert A. Amezquita (aut, cre, cph): robert.amezquita@fredhutch.org