SV survey

pub

caller class

input type

somatic/denovo

from

validated vs.

cited

used by

compared by

algorithm

features

description

installation

study

source

SvABA

2017

low memory, fast

http://biorxiv.org/content/early/2017/02/01/105080

valor

2017

long range sequencing e.g. 10X Genomics linked-read sequencing, pooled clone sequencing

https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-016-3444-1

novobreak

2017

assembly, md anderson

http://www.nature.com/nmeth/journal/v14/n1/full/nmeth.4084.html

paywall

VarDict

2016

SV/indel/SNV/LOH

targeted bam

somatic

Astra Zeneca

GATK UG/HC, Freebayes, varscan, pindel, scalpel, manta, lumpy

SomaticSeq, RAVE, bcbio

Bcb8

consensus on realigned soft-clipped reads used as search query

Efficient with ultra deep seq. calls complex variants (same read, multiple vars). filters PCR artifacts. Estimates SV allele frequency. Large insertions not yet called. inter-chromosomal fusion not yet called. can perform amplicon-aware calling but for single samples only.

Calls SNV, MNV, InDels, complex and structural variants, performs local realignments on the fly. Performance scales linearly to sequencing depth. Performs amplicon aware variant calling for polymerase chain reaction (PCR)-based targeted sequencing often used in diagnostic settings. Is able to detect PCR artifacts. Detects differences in somatic and loss of heterozygosity variants between paired samples.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4914105/

https://github.com/AstraZeneca-NGS/VarDict

gridss

2016

poster

bcbio

de bruijn

http://smbe-2016.p.asnevents.com.au/days/2016-07-05/abstract/35482

https://github.com/PapenfussLab/gridss/

SV-Bay

2016

paired-end, mate pair

somatic

Curie Institute

gasvpro, breakdancer, lumpy, delly

PEM & read coverage with bayesian testing for adjacencey

works with paired end or mate pair; analyzes tumor/normal pair concurrently

combine PEM signatures and DOC lanking each candidate rearrangement. takes into account GC-content and mappability. Bayesian framework based on both PEM and DOC information ; infers 15 different types of structural variant from the detected novel genomic adjacencies

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4896370/

https://github.com/InstitutCurie/SV-Bay

COSMOS

2016

somatic

Advanced Industrial Science and Technology (AIST), Tokyo

Breakdancer, GasVPro, Delly, Lumpy, first using synthetic SVs introduced to hg19 with simulated impurity levels, then using hypermutable mouse ESC model with gamma irradiation

discordant pair reads, classify, DOC binomial

fast. mouse model for validation plus synthetic

compares the statistics of the mapped read pairs in tumor samples with isogenic normal control samples in a distinct asymmetric manner.

http://nar.oxfordjournals.org/content/44/8/e78

http://wall-lab.stanford.edu/projects/cosmos/

svstat

2016

https://scfbm.biomedcentral.com/articles/10.1186/s13029-016-0051-0

svelter

2016

complex rearrangements

https://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0993-1

skald

2016

http://www.nature.com/articles/npjgenmed201626?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+npjgenmed%2Frss%2Fcurrent+(npj+Genomic+Medicine)

devro

2016

population: paired end/depth of coverage

http://biorxiv.org/content/early/2016/12/15/094474

seq2c

2015

n/a

https://github.com/AstraZeneca-NGS/Seq2C

Manta

2015

SV/indel

wg bam trios

germ

Illumina

pindel, delly

bcbio, bioconda

breakend graph/assembly

Parallelized for clusters. handles degraded FFPE samples, uses pedigree-consistency and cosmic for validation. way faster than delly

provides scoring models for germline analysis of diploid individuals and somatic analysis of tumor-normal sample pairs, with additional applications under development for RNA-Seq, de novo variants, and unmatched tumors. less than a tenth of the time that comparable methods require

https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btv710

https://github.com/Illumina/manta

MetaSV

2015

Bina/Roche

pindel, BreakSeq2, LUMPY, BreakDancer, Delly, CNVNator, MindTheGap

bioconda, bcbio (http://blog.bina.com/read/metasv-integration-in-bcbio)

merge then assemble: pindel, BreakSeq2, LUMPY, BreakDancer, Delly, CNVNator, MindTheGap

consensus caller

merging SVs from multiple tools for all types of SVs. It also analyzes soft-clipped reads from alignment to detect insertions accurately since existing tools underestimate insertion SVs. Local assembly in combination with dynamic programming is used to improve breakpoint resolution. Paired-end and coverage information is used to predict SV genotypes.

https://www.ncbi.nlm.nih.gov/pubmed/25861968

http://bioinform.github.io/metasv/

Wham

2015

denovo

U of Utah

lumpy, delly, softsearch

bcbio, biocondor

bcb8

mate-pair & split read mapping, soft-clipping, alternative alignment, consensus sequence based evidence

requires bwa. does association testing

pinpoint SVs in pooled and genotypic data associated with phenotypic variation. uses split-read, mate-pair, and alternative alignments to find the other SV breakpoint. Positions in the pileup where three or more primary reads share the same breakpoint are interrogated as a putative SV. Use SA and XA cigar tags as alternative alignment locations, cluster those. SW align clipped consensus to alternative locations. intra-chromosomal require min 2 reads.

http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004572

https://github.com/zeeev/wham

BreakMer

2015

targeted seq

somatic

Broad, MacConaill

crest, meerkat, breakdancer, pindel

soft clip, identify kmers in reads but not in reference

uses a ‘kmer’ strategy to assemble misaligned sequence reads for predicting insertions, deletions, inversions, tandem duplications and translocations at base-pair resolution in targeted resequencing data. Variants are predicted by realigning an assembled consensus sequence created from sequence reads that were abnormally aligned to the reference genome

http://nar.oxfordjournals.org/content/43/3/e19

https://github.com/ccgd-profile/BreaKmer

indelMiner

2015

indel

wg bam pair

simple de novo, somatic

Penn State U

samtools, pindel, prism

split-read, paired-end, soft-clipped. align unmapped reads at both ends to look for indels

validatin against synthetic variants introduced to chr22 and the na18507 data set. recommended to align with gatk indelRealigner

uses a split-read approach to identify the precise breakpoints for indels of size less than a user specified threshold, and supplements that with a paired-end approach to identify larger variants that are frequently missed with the split-read approach

http://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-015-0483-6

https://github.com/aakrosh/indelMINER

BreakSeek

2015

Chinese Academy of Sciences

pindel, lumpy, crest, soapindel, breakdancer, prism, delly

soft-clipping/breakread break points, paired-end span to validate indel, sophisticated probabilistic scoring model

describes parameters for competitors, works reasonable well for all size indels, estimates level of heterozygosity

unbiasedly and efficiently detect both homozygous and heterozygous INDELs, ranging from several base pairs to over thousands of base pairs, with accurate breakpoint and heterozygosity rate estimations

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4538813/

https://sourceforge.net/projects/breakseek/

RAPTR-SV

2015

USDA

discordant read-pair, split read, soft-clip, filter

sensitivity for tandem duplications

combining their predictions to generate highly confident SV calls, which can be filtered at runtime for improved accuracy.

https://academic.oup.com/bioinformatics/article/31/13/2084/195773/RAPTR-SV-a-hybrid-method-for-the-detection-of

https://github.com/njdbickhart/RAPTR-SV

speedseq

2015

WashU St Louis Hall

meta. freebayes, lumpy, cnvnator, and custom caller svtyper.

assumes diploid. uses pedigree+ to call validation variants. discusses parameters. reports confidence score.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4589466/

https://github.com/hall-lab/speedseq

Lumpy

2014

U Virginia

gasvpro, delly, pindel

104

bcbio, metasv, biocondor

bcb8, bcb3

merge read-pair, split read, read-depth in a breakpoint probability map. classify and cluster

sensitive to low MAF. not great for small dels

LUMPY integrates disparate signals by converting them to a common format in which the two predicted breakpoint intervals in the reference genome are represented as paired probability distributions.

https://genomebiology.biomedcentral.com/articles/10.1186/gb-2014-15-6-r84

https://github.com/arq5x/lumpy-sv

Scalpel

2014

indel

exome-capture

denovo

Cold Spring Harbor, Simons Center for Quantitative Biology

gatk hc, SOAPindel

somaticseq, bcbio, biocondor

Bcb8

de bruijn graph traversal, local assembly with iterative k-mer k value reassessment to eliminate repeats.

slow, not for wgs. does indel normalization

localized micro-assembly of specific regions of interest with the goal of detecting mutations with high accuracy and increased power. It is based on the de Bruijn graph assembly paradigm and implements an on-the-fly repeat composition analysis coupled with a self-tuning k-mer strategy

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4180789/

http://scalpel.sourceforge.net/

gindel

2014

indels > 50bp

U Conneticut

pindel, cleversv

SVM on 7 features: discordant pair, split-read, read depth, concordant encompassing pair, single-end-mapped pair, partially mapped reads, fully-mapped spanning reads,

efficient

An approach for calling genotypes of both insertions and deletions from sequence reads. GINDEL uses a machine learning approach which combines multiple features extracted from next generation sequencing data. It performs well for insertion genotyping on both simulated and real data. GINDEL can not only call genotypes of insertions and deletions (both short and long) for high and low coverage population sequence data, but also is more accurate and efficient than other approaches.

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0113324

https://sourceforge.net/projects/gindel/

Gustaf

2014

Freie U, Berlin

delly, pindel

biocondor

local alignment between unmapped reads shows breakpoints

small validation set. claims best for small SVs 30-100bp. small validation set. might work with FFPE

based on a generic multi-split alignment strategy that can identify SV breakpoints with base pair resolution.

http://bioinformatics.oxfordjournals.org/content/30/24/3484

http://www.seqan.de/apps/gustaf/

SMuFin

2014

SV/SNV

paired exome fastqs

somatic

Barcelona Supercomputing Center

mutect, breakdancer, pindel, delly, crest

directly compares reads "quaternary sequence tree"

exome only, too slow for WG. parallelized

directly compares sequence reads from normal and tumor genomes to accurately identify and characterize a range of somatic sequence variation, from single-nucleotide variants (SNV) to large structural variants at base pair resolution.

http://www.nature.com/nbt/journal/v32/n11/full/nbt.3027.html

http://cg.bsc.es/smufin/

Socrates

2014

somatic

The Walter and Eliza Hall Institute of Medical Research

re-aligns and clusters soft-clipped reads

needs parameterized. fast. split-read only algorithms are better for short read FFPE data. "On real tumour data without additional information, we find it impractical to run at its most sensitive settings, but it is easily tuned."

uses split reads to find breakpoints. It is optimized to be fast and extremely sensitive.

http://bioinformatics.oxfordjournals.org/content/30/8/1064

https://github.com/PapenfussLab/socrates

Ulysses

2014

mate-pair only

Lab of Computational and Quantitative Biology, Paris

assessing, in a principled manner, the statistical significance of each possible variant (duplications, deletions, translocations, insertions and inversions) against an explicit model for the generation of experimental noise.

http://bioinformatics.oxfordjournals.org/content/31/6/801.full?keytype=ref&%2520ijkey=fiLPQEO731TMaNt

http://www.lcqb.upmc.fr/ulysses/#citeUlysses

ViVar

2014

Ghent U, Belgium

needs a reference set for sequencing error model

facilitates the processing, analysis and visualization, of structural variation based on massive parallel sequencing data

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4264741/

https://www.cmgg.be/vivar/

Bellerophon

2013

SV(translocs)

Case Western

GASV, Breakdancer, SVDetect, and CREST

discordant reads, soft-clipped

specific to interchromosomal translocations. really basic caller

uses discordant read pairs and "soft-clipped" reads to predict the location of the precise breakpoints. for each chimeric breakpoint, attempts to classify it as a participant in an unbalanced translocation, balanced translocation, or interchromosomal insertion.

https://www.ncbi.nlm.nih.gov/pubmed/23734783

http://cbc.case.edu/Bellerophon/

sv-m

2013

https://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-14-132

pesv-fisher

2013

pem & doc

https://www.ncbi.nlm.nih.gov/pubmed/23704902/

iSVP

2013

breakdancer, delly, pindell, haplotypecaller

limit each method's calls to their optimal size ranges

meta. deletions only

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4029547/

Meerkat

2013

somatic

Harvard, Park

n/a

110

discordant read pair clustering with refinement

may recognize a relatively narrow scope of unique variants from specific complex rearrangement events like dna repair pathways. identified hundreds of new complex variants in na12878 and na18507. probably a good addition to any meta caller. d

considers local clusters of discordant read pairs to recognize specific complex events. uses split, clipped, and multiple-aligned reads

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3704973/

http://compbio.med.harvard.edu/Meerkat/

soapindel

2013

indels

BGI Shenzhen

dindel, pindel, gatk

identifies breakpoints from discordant reads, multi-path de bruijn graph assembly

similar sensitivity and specificity for small indels, higher sensitivity for large indels. might be slow. should call SNPs too. assigns confidence q-scores. weird validation looking at hg19 vs venter genome and chimpanzee vs hg19

assign all unmapped reads with a mapped partner to their expected genomic positions and then perform extensive de novo assembly on the regions with many unmapped reads to resolve homozygous, heterozygous, and complex indels by exhaustive traversal of the de Bruijn graph

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3530679/

http://soap.genomics.org.cn/soapindel.html

SoftSearch

2013

Mayo Clinic, Kocher

breakdancer, delly, crest, svseq

soft-clipping heuristic, number of soft-clipped reads per position

slow but high TP rate. works at low depth but needs parameters adjusted. Levenstein Distance "confidence scores"

Assuming soft clipping delineates the exact breakpoint position and direction, DRPs overlapping such soft-clipped areas should already contain the information about the type and size of SV, obviating the need for secondary alignments.

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0083356

http://bioinformaticstools.mayo.edu/research/softsearch/

TIGRA

2013

breakpoints

population locate common alleles. can be used as de novo?

WashU MD Anderson Cancer Center, Weinstock

spades, velvet, sga, phrap

iterative breakpoint collection, de bruijn graph, assembles the breakpoints

uses population data, low FDR

http://genome.cshlp.org/content/early/2013/12/04/gr.162883.113

http://bioinformatics.mdanderson.org/main/TIGRA

Delly

2012

SV/CNVs

EMBL

pindel, breakdancer, gasv, hydra

273

bcbio, metasv, biocondor

bcb8, cbc3

integrates short insert paired-ends and long-range mate-pairs to identify discordant pairs, then uses split-read alignments to identify breakpoints

high sensitivity and specificity, lower sensitivity to small deletions

integrates short insert paired-ends, long-range matepairs and split-read alignments to accurately delineate genomic rearrangements

http://bioinformatics.oxfordjournals.org/content/28/18/i333.abstract

https://github.com/dellytools/delly

cn.MOPS

2012

CNV

population

Johannes Kepler U, Australia

133

biocondor

bcb3

decomposes variations in the depth of coverage across samples into integer copy numbers and noise by means of its mixture components and Poisson distributions

http://nar.oxfordjournals.org/content/40/9/e69

http://www.bioconductor.org/packages/release/bioc/html/cn.mops.html

cpgbattenberg

2012

608

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3428864/

BreakPointer

2012

Max Plank Institute, Haas

pindel

single-end read breakpoint locator

By taking advantage of local non-uniform read distribution and misalignments created by SVs, Breakpointer scans the alignment of single-end reads to identify regions containing potential breakpoints.

http://bioinformatics.oxfordjournals.org/content/28/7/1024

https://github.com/ruping/Breakpointer

CLEVER

2012

Life Sciences Group, Amsterdam

gasv, variationhunter, breakdancer, hydra

biocondor

clustering on concordant pairs

better at 20-100bp size range

enumerates all max-cliques and statistically evaluates them for their potential to reflect insertions or deletions.

http://bioinformatics.oxfordjournals.org/content/28/22/2875.long

http://www.mybiosoftware.com/clever-2-0rc1-clique-enumerating-variant-finder.html

ForestSV

2012

random forests

https://www.ncbi.nlm.nih.gov/pubmed/22751202

http://sebatlab.ucsd.edu/index.php/software-data

GasVPro

2012

Brown U

hydra, breakdancer, CNVer

joint P on paired read and read depth

validates vs HuRef, NA18705, and NA12878, has ROC curves. models uncertainty in call/reference overlap for truth calls to be more precise

Combines read depth information along with discordant paired-read mappings into a single probabilistic model two common signals of structural variation.

http://genomebiology.biomedcentral.com/articles/10.1186/gb-2012-13-3-r22

https://code.google.com/archive/p/gasv/

Prism

2012

U Toronto,

pindel, svseq, splitread, breakdancer, CNVnator, crest

uses a split-alignment approach informed by the mapping of paired-end reads, hence enabling breakpoint identification of multiple SV types, including arbitrary-sized inversions, deletions and tandem duplications

http://bioinformatics.oxfordjournals.org/content/28/20/2576

http://compbio.cs.toronto.edu/prism/

SplitRead

2012

exome

de novo

Howard Hughs Medical Institute

discordant pairs clustering, map with mrsFast and hamming distance, call anomolous mappings, split unmapped reads, search.

compares read depth between parents and child to identify de novo mutations

searches for clusters of mate pairs where one end maps to the reference genome but the other end does not because it traverses a breakpoint creating a mapping inconsistency with respect to the reference sequence

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3269549/

http://splitread.sourceforge.net/

ClipCrop

2011

U Tokyo

breakdancer, cnvnator, pindel

not for somatic. doesn't recognize useful mutation types according to socrates authors

A soft-clipped sequence is an unmatched fragment in a partially mapped read

http://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-12-S14-S7

https://github.com/shinout/clipcrop

CREST

2011

somatic

St Jude, Zhang

260

single read soft clipping, classification

made for somatic comparisons, lower performance for small deletions

uses the soft-clipping reads to directly map the breakpoints of structural variations

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3527068/

http://www.stjuderesearch.org/site/lab/zhang

GenomeStrip 1 & 2

2011

SV/CNVs

bams

population

Broad, McCarroll

spanner, pindel, breakdancer, pemer, cnvnator

206

discordant read pair clustering, reassemble breakpoint-spanning reads by allele, read-depth for copy number estimate, align unmapped reads to breakpoint database

validated by 1kGP, "most sensitive and accurate". less sensitive to small SVs. 2.0 adds CNV detection

designed to find shared variation using data from multiple individuals. Genome STRiP looks both across and within a set of sequenced genomes to detect variation.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5094049/

http://software.broadinstitute.org/software/genomestrip/

inGap-SV

2011

Chinese Academy of Sci, Zhao

Breakdancer, variationhunter, spanner, PEMer, cortex, pindel

paired-end mapping & depth of coverage

good validation against 12878, differentiates homo- and hetero-zygous variants

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3125812/

http://ingap.sourceforge.net/

hydra

2010

sanger split-read & illumina paired-end

U Va

none

195

split-read + paired end

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2860164/

https://code.google.com/archive/p/hydra-sv/

AGE

2010

Yale, Gerstein

biocondor

read-depth

improved alignment algorithm, but does not call variants. narrow scope validation, proof of theory. has been modified for metasv inclusion

AGE for Alignment with Gap Excision, finds the optimal solution by simultaneously aligning the 5′ and 3′ ends of two given sequences and introducing a ‘large-gap jump’ between the local end alignments to maximize the total alignment score. We also describe extensions allowing the application of AGE to tandem duplications, inversions and complex events involving two large gaps.

http://bioinformatics.oxfordjournals.org/content/27/5/595.abstract

http://sv.gersteinlab.org/age/

slope

2010

WashU St Louis, Pfiefer

pindel, breakdancer

basic simulated data

detect sequence breakpoints from only one side of a split read, and therefore does not rely on the insert size for detection.

https://academic.oup.com/bioinformatics/article/26/21/2684/214667/SLOPE-a-quick-and-accurate-method-for-locating-non

http://www-genepi.med.utah.edu/suppl/SLOPE/slope_guide.txt

SVDetect

2010

paired-end/mate pairs

Curie Institute

GasV

131

discordant read pairs, adds mate-pairs, sliding window for clustering

anomalously mapped read pairs provided by current short read aligners to localize genomic rearrangements and classify them according to their type, e.g. large insertions– deletions, inversions, duplications and balanced or unbalanced interchromosomal translocations.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2905550/

http://svdetect.sourceforge.net/Site/Manual.html

SVMerge

2010

Sanger Institute

meta - BDMax, Pindel, SECluster, RetroSeq, RDXplorer

http://genomebiology.biomedcentral.com/articles/10.1186/gb-2010-11-12-r128

http://svmerge.sourceforge.net//

Pindel

2009

EMBL, Ning

768