In PapenfussLab/RTDetect: Retrotransposed transcript detection from structural variants

options(width=1000)
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-"
)

svaRetro: R package for retrotransposed transcript detection from structural variants

svaRetro contains functions for detecting retrotransposed transcripts (RTs) from structural variant calls. It takes structural variant calls in GRanges of breakend notation and identifies RTs by exon-exon junctions and insertion sites. The candidate RTs are reported by events and annotated with information of the inserted transcripts.

This package uses a breakend-centric event notation adopted from the StructuralVariantAnnotation package. 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").

Installation

svaNUMT is currently available for download in Bioconductor (since BioC 3.14 & R 4.1):

# install.packages("BiocManager")
BiocManager::install("svaRetro")

The development version can be installed from GitHub:

BiocManager::install("PapenfussLab/svaRetro")

How to cite

If you use svaRetro, please cite svaRetro here.

@article {Dong2021.08.18.456578,
    author = {Dong, Ruining and Cameron, Daniel and Bedo, Justin and Papenfuss, Anthony T},
    title = {svaRetro and svaNUMT: Modular packages for annotation of retrotransposed transcripts and nuclear integration of mitochondrial DNA in genome sequencing data},
    elocation-id = {2021.08.18.456578},
    year = {2021},
    doi = {10.1101/2021.08.18.456578},
    publisher = {Cold Spring Harbor Laboratory},
    abstract = {Background The biological significance of structural variation is now more widely recognized. However, due to the lack of available tools for downstream analysis, including processing and annotating, interpretation of structural variant calls remains a challenge.Findings Here we present svaRetro and svaNUMT, R packages that provide functions for annotating novel genomic events such as non-reference retro-copied transcripts and nuclear integration of mitochondrial DNA. We evaluate the performance of these packages to detect events using simulations and public benchmarking datasets, and annotate processed transcripts in a public structural variant database.Conclusions svaRetro and svaNUMT provide efficient, modular tools for downstream identification and annotation of structural variant calls.Competing Interest StatementThe authors have declared no competing interest.SVstructural variantNUMTnuclear mitochondrial integrationRTretroposed transcriptTSDtarget site duplicationmtDNAmitochondrial DNA},
    URL = {https://www.biorxiv.org/content/early/2021/08/19/2021.08.18.456578},
    eprint = {https://www.biorxiv.org/content/early/2021/08/19/2021.08.18.456578.full.pdf},
    journal = {bioRxiv}
}

Workflow

Below is a workflow example for detecting RTs from a human SV callset. This example is taken from the vignette of svaRetro.

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_vcf <- readVcf(system.file("extdata", "diploidSV.vcf", package = "svaRetro"))
RT_gr <- StructuralVariantAnnotation::breakpointRanges(RT_vcf, nominalPosition=TRUE)
RT <- rtDetect(RT_gr, hg19.genes, maxgap=50, minscore=0.3)

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()

PapenfussLab/RTDetect documentation built on June 15, 2022, 1:42 a.m.