ciRcus: an R package for circRNA manipulation, annotation and analysis

ciRcus is a collection of functions for everyday munging of circRNA data. In its current, preliminary version, it can take lists of putative splice junctions generated using find_circ.py (Memczak et al. 2013, circBase) as input, and perform following annotation steps:

• quality filtering
• suggest a host gene a circRNA candidate was spliced from
• calculate circular-to-linear ratio
• describe gene features a circRNA candidate was spliced from
• report if a circRNA's splice junctions are already annotated as linear exon-intron junctions
• generate read count histogram and gene features pie-charts

Installation

Install via install_github()

#' Install dependecies
install.packages( c("data.table", "DBI", "hash", "ggplot2", "RMySQL", "devtools"))
source("http://bioconductor.org/biocLite.R")
biocLite(c("GenomicRanges","GenomicFeatures", "IRanges", "biomaRt", "AnnotationHub"))

#' install the package
library(devtools)
install_github("BIMSBbioinfo/ciRcus", build_vignettes=FALSE)

Using the package

Build TxDb object with genomic features and save locally

Load genomic features from Ensembl and build a database for later (re)use. Currently supported assemblies are hg19, hg38, mm10, rn5, dm6 and WBcel235. This needs to be done only once per assembly.

gtf2sqlite( assembly = "hg19",
            db.file  = system.file("extdata/db/human_hg19_ens75_txdb.sqlite",
                                   package="ciRcus"))

Extract features from the database

List of features returned by loadAnnotation() will be used to annotate circRNAs. Saving it as a separate object is a good practice once we start analyzing multiple circRNA libraries.

annot.list <- loadAnnotation(system.file("extdata/db/human_hg19_ens75_txdb.sqlite",
                                         package="ciRcus"))

Load and annotate circRNAs

cdata <- data.frame(sample=c("FC1", "FC2", "H1", "H2", "L1", "L2"),
                    filename=list.files(system.file('extdata/encode_demo_small', package='ciRcus'),
                                        pattern='sites.bed',
                                        full.names=TRUE)[1:6])
circs.se <- summarizeCircs(colData=cdata, wobble=1, keepCols=1:12)
circs.se <- annotateCircs(circs.se, annot.list=annot.list)
circs.dt <- resTable(circs.se)
circs.dt

Plot data

histogram(circs.se, 0.5)
annotPie(circs.se, 0.02)
uniqReadsQC(circs.se, "all")