Importing tRNAscan-SE output as GRanges

BiocStyle::markdown(css.files = c('custom.css'))

Introduction

tRNAscan-SE [@Lowe.1997] can be used for prediction of tRNA genes in whole genomes based on sequence context and calculated structural features. Many tRNA annotations in genomes contain or are based on information generated by tRNAscan-SE, for example the current SGD reference genome sacCer3 for Saccharomyces cerevisiae. However, not all available information from tRNAscan-SE end up in the genome annotation. Among these are for example structural information, additional scores and the information, whether the conserved CCA-end is encoded in the genomic DNA. To work with this complete set of information, the tRNAscan-SE output can be parsed into a more accessible GRanges object using tRNAscanImport.

Getting started

The default tRNAscan-SE output, either from running tRNAscan-SE [@Lowe.1997] locally or retrieving the output from the gtRNADb [@Chan.2016], consist of a formatted text document containing individual text blocks per tRNA delimited by an empty line.

suppressPackageStartupMessages({
  library(tRNAscanImport)
})
library(tRNAscanImport)
yeast_file <- system.file("extdata",
                          file = "yeast.tRNAscan",
                          package = "tRNAscanImport")

# output for sacCer3
# Before
readLines(con = yeast_file, n = 7L)

Importing as GRanges

To access the information in a BioC context the import as a GRanges object comes to mind. import.tRNAscanAsGRanges() performs this task by evaluating each text block using regular expressions.

# output for sacCer3
# After
gr <- import.tRNAscanAsGRanges(yeast_file)
head(gr, 2)
# Any GRanges passing this, can be used for subsequent function
istRNAscanGRanges(gr)

The result can be used directly in R or saved as gff3/fasta file for further use, including processing the sequences for HTS read mapping or statistical analysis on tRNA content of the analyzed genome.

suppressPackageStartupMessages({
  library(Biostrings)
  library(rtracklayer)
})
library(Biostrings)
library(rtracklayer)
# suppressMessages(library(rtracklayer, quietly = TRUE))
# Save tRNA sequences
writeXStringSet(gr$tRNA_seq, filepath = tempfile())
# to be GFF3 compliant use tRNAscan2GFF
gff <- tRNAscan2GFF(gr)
export.gff3(gff, con = tempfile())

Visualization

The tRNAscan-SE information can be visualized using the gettRNAFeaturePlots() function of the tRNA package, returning a named list of ggplot2 plots, which can be plotted or further modified. Alternatively, gettRNASummary() returns the aggregated information for further use.

# tRNAscan-SE output for hg38
human_file <- system.file("extdata",
                          file = "human.tRNAscan",
                          package = "tRNAscanImport")
# tRNAscan-SE output for E. coli MG1655
eco_file <- system.file("extdata",
                        file = "ecoli.tRNAscan",
                        package = "tRNAscanImport")
# import tRNAscan-SE files
gr_human <- import.tRNAscanAsGRanges(human_file)
gr_eco <- import.tRNAscanAsGRanges(eco_file)

# get summary plots
grl <- GRangesList(Sce = gr,
                   Hsa = gr_human,
                   Eco = gr_eco)
plots <- gettRNAFeaturePlots(grl)
plots$length
plots$tRNAscan_score
plots$gc
plots$tRNAscan_intron
plots$variableLoop_length

Further reading

Further examples of working with tRNA information can be found in the vignette of the tRNA package.

Session info

sessionInfo()

References



Try the tRNAscanImport package in your browser

Any scripts or data that you put into this service are public.

tRNAscanImport documentation built on Nov. 1, 2018, 3:30 a.m.