In PapenfussLab/RTDetect: Retrotransposed transcript detection from structural variants

options(width=1000)
knitr::opts_chunk$set(
  collapse = TRUE, 
  comment = "#>")

Introduction

This vignette outlines a workflow of detecting retrotransposed transcripts (RTs) from Variant Call Format (VCF) using the svaRetro package.

Installation

The svaRetro package can be installed from Bioconductor as follows:

if(!requireNamespace("BiocManager", quietly = TRUE))
  install.packages("BiocManager")
BiocManager::install("svaRetro")

Using GRanges for structural variants: a breakend-centric data structure

This package uses a breakend-centric event notation adopted from the StructuralVariantAnnotation package. In short, breakends are stored in a GRanges object with strand used to indicate breakpoint orientation, where breakpoints are represented using a partner field containing the name of the breakend at the other side of the breakend. This notation was chosen as it simplifies the annotations of RTs which are detected at breakend-level.

Workflow

Loading data from VCF

VCF data is parsed into a VCF object using the readVCF function from the Bioconductor package VariantAnnotation. Simple filters could be applied to a VCF object to remove unwanted calls. The VCF object is then converted to a GRanges object with breakend-centric notations using StructuralVariantAnnotation. More information about VCF objects and breakend-centric GRanges object can be found by consulting the vignettes in the corresponding packages with browseVignettes("VariantAnnotation") and browseVignettes("StructuralVariantAnnotation").

library(StructuralVariantAnnotation)
library(VariantAnnotation)
library(svaRetro)

RT_vcf <- readVcf(system.file("extdata", "diploidSV.vcf", package = "svaRetro"))

RT_gr <- StructuralVariantAnnotation::breakpointRanges(RT_vcf, 
                                                       nominalPosition=TRUE)
head(RT_gr)

Note that StructuralVariantAnnotation requires the GRanges object to be composed entirely of valid breakpoints. Please consult the vignette of the StructuralVariantAnnotation package for ensuring breakpoint consistency.

Identifying Retrotransposed Transcripts

The package provides rtDetect to identify RTs using the provided SV calls. This is achieved by detecting intronic deletions, which are breakpoints at exon-intron (and intron-exon) boundaries of a transcript. Fusions consisting of an exon boundary and a second genomic location are reported as potential insertion sites. Due to the complexity of RT events, insertion sites can be discovered on both left and right sides, only one side, or none at all.

library(TxDb.Hsapiens.UCSC.hg19.knownGene)
library(dplyr)
hg19.genes <- TxDb.Hsapiens.UCSC.hg19.knownGene

RT <- rtDetect(RT_gr, hg19.genes, maxgap=10, minscore=0.8)

The output is a list of GRanges object consisting of two sets of GRanges calls, insSite and junctions, containing candidate insertion sites and exon-exon junctions respectively. Candidate insertion sites are annotated by the source transcripts and whether exon-exon junctions are detected for the source transcripts. RT junction breakends are annotated by the UCSC exon IDs, corresponding transcripts, and NCBI gene symbols.

RT$SKA3

Visualising breakpoint pairs via circos plots

One way of visualising RT is by circos plots. Here we use the package circlize to demonstrate the visualisation of insertion site and exon-exon junctions.

To generate a simple circos plot of RT event with SKA3 transcript:

library(circlize)
rt_chr_prefix <- c(RT$SKA3$junctions, RT$SKA3$insSite)
seqlevelsStyle(rt_chr_prefix) <- "UCSC"
pairs <- breakpointgr2pairs(rt_chr_prefix)
pairs

To see supporting breakpoints clearly, we generate the circos plot according to the loci of event.

circos.initializeWithIdeogram(
    data.frame(V1=c("chr13", "chr11"),
               V2=c(21720000,108585000),
               V3=c(21755000,108586000),
               V4=c("q12.11","q24.3"),
               V5=c("gneg","gpos50")))
circos.genomicLink(as.data.frame(S4Vectors::first(pairs)), 
                   as.data.frame(S4Vectors::second(pairs)))
circos.clear()