README.md

DOI

FantomTSS.hg19

The goal of this package is to easily extract the genomic positions and the normalized expression of Fantom's TSS (in TPM) in hg19.

Installation

require(devtools)
devtools::install_github("CharlesJB/FantomTSS.hg19")

Main functions

The get_fantom_tss functions returns a GRanges object with all the tss and no metadata columns:

get_fantom_tss()

The get_fantom_tss_tpm returns a GRanges object with all the tss and selected metadata columns:

# To get the expression of tss in A549
get_fantom_tss_tpm(cell_lines = "A549")

# To get the expression of tss in A549 and K562
get_fantom_tss_tpm(cell_lines = c("A549", "K562"))

# To merge all the columns for a cell line by calculating the mean value for
# each tss
get_fantom_tss_tpm(cell_lines = c("A549", "K562"), merge.FUN = mean)

# To merge all the columns for a cell line by calculating the sum for each
# tss
get_fantom_tss_tpm(cell_lines = c("A549", "K562"), merge.FUN = sum)

Reproduce the datasets

If you wish to reproduce the datasets, you can use the prepare_datasets function. From a R session in the packages main directory:

library(devtools)
library(data.table)
load_all()
source("data-raw/data.R")
prepare_datasets()

Note that this will download large files (~800 MB) from Fantom's web page. You will also need to reinstall the package if you want to use those new datasets.



CharlesJB/FantomTSS.hg19 documentation built on May 6, 2019, 9:57 a.m.