BSgenome-automatic-forcing

BSgenome-automatic-forcing is a script to automatically force a BSgenome data package. Currently, this script supports you to forge a package from Ensembl fasta files (HomoSapiens - release 37.75). For futher modifications, referring to next section

Requirements

Obviously, you need to install BSgenome pakage first.

Installation

Simply executing bash install.sh on the terminal. HomoSapiens(37.75) fasta files from Ensembl will be downloaded and BSgenome.Sapiens.Ensembl.3775 package will be built and installed on your environment.

Warning: If .TwoBits_export error happens, please create extdata following the author's suggestion:

cd `echo 'cat(system.file(package="BSgenome"))' | R --vanilla --slave`
cd pkgtemplates/BSgenome_datapkg/
mkdir inst
mkdir inst/extdata

Example

After executing the script, you can try an example as follows:

> library(BSgenome.Sapiens.Ensembl.3775)
> Sapiens[["1"]]

  249250621-letter "DNAString" instance
seq: NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN...NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN

> gr = GRanges(seqnames=c("1", "2", "MT"), ranges=IRanges(start=101:103, width=9))
> getSeq(Sapiens, gr)

  A DNAStringSet instance of length 3
    width seq
[1]     9 NNNNNNNNN
[2]     9 NNNNNNNNN
[3]     9 GCCGGAGCA

How to modify

1. Download paths: You can change the link pattern in wget command or simply remove these lines if you already have your own fasta files in seq folder.
2. seed_template: If you use your own dataset, you have to modify this file and be careful for some parameters as follows:

• Package: name of the package you will use to call library.
• BSgenomeObjname: short name of the genome, e.g. Sapiens.
• seqnames: (vector) name of fasta files in seq folder, also name of chromosomes.