svviz: a read viewer for validating structural variants

FuSViz-visualization and interpretation of structural variation using cancer genomics and transcriptomics data

Structural variation (SV) is a frequent category of genetic alterations important for understanding cancer genome evolution and revealing key cancer driver events. With the development of high-throughput sequencing technologies, the ability to detect SVs of various sizes and types has improved, at both the DNA and RNA levels. However, SV calls are still prone to a considerable fraction of false positives, which necessitates visual inspection and manual curation as part of the quality control process. Identification of reliable and recurrent SVs in larger cohorts lends strength to revealing the driving roles of SVs in cancer development and to the discovery of potential diagnostic and prognostic biomarkers. Here, we present FuSViz, an application for visualization, interpretation, and prioritization of SVs. The tool provides multiple data view approaches in a user-friendly interface, allowing the investigation of prevalence and recurrence of SVs and relevant partner genes in a sample cohort. It integrates SV calls from DNA and RNA sequencing datasets to comprehensively illustrate the biological impact of SVs on the implicated genes and associated genomic regions. The functionality of FuSViz is intended for interrogation of both recurrent and private SVs, effectively assisting with pathogenicity evaluation and biomarker discovery in cancer sequencing projects.

NPSV-deep: a deep learning method for genotyping structural variants in short read genome sequencing data

MOTIVATION

Structural variants (SVs) play a causal role in numerous diseases but can be difficult to detect and accurately genotype (determine zygosity) with short-read genome sequencing data (SRS). Improving SV genotyping accuracy in SRS data, particularly for the many SVs first detected with long-read sequencing, will improve our understanding of genetic variation.

RESULTS

NPSV-deep is a deep learning-based approach for genotyping previously reported insertion and deletion SVs that recasts this task as an image similarity problem. NPSV-deep predicts the SV genotype based on the similarity between pileup images generated from the actual SRS data and matching SRS simulations. We show that NPSV-deep consistently matches or improves upon the state-of-the-art for SV genotyping accuracy across different SV call sets, samples and variant types, including a 25% reduction in genotyping errors for the Genome-in-a-Bottle (GIAB) high-confidence SVs. NPSV-deep is not limited to the SVs as described; it improves deletion genotyping concordance a further 1.5 percentage points for GIAB SVs (92%) by automatically correcting imprecise/incorrectly described SVs.

AVAILABILITY AND IMPLEMENTATION

Python/C++ source code and pre-trained models freely available at https://github.com/mlinderm/npsv2.

Investigating structural variant, indel and single nucleotide polymorphism differentiation between locally adapted Atlantic salmon populations

Genomic structural variants (SVs) are now recognized as an integral component of intraspecific polymorphism and are known to contribute to evolutionary processes in various organisms. However, they are inherently difficult to detect and genotype from readily available short-read sequencing data, and therefore remain poorly documented in wild populations. Salmonid species displaying strong interpopulation variability in both life history traits and habitat characteristics, such as Atlantic salmon (Salmo salar), offer a prime context for studying adaptive polymorphism, but the contribution of SVs to fine-scale local adaptation has yet to be explored. Here, we performed a comparative analysis of SVs, single nucleotide polymorphisms (SNPs) and small indels (<50 bp) segregating in the Romaine and Puyjalon salmon, two putatively locally adapted populations inhabiting neighboring rivers (Québec, Canada) and showing pronounced variation in life history traits, namely growth, fecundity, and age at maturity and smoltification. We first catalogued polymorphism using a hybrid SV characterization approach pairing both short- (16X) and long-read sequencing (20X) for variant discovery with graph-based genotyping of SVs across 60 salmon genomes, along with characterization of SNPs and small indels from short reads. We thus identified 115,907 SVs, 8,777,832 SNPs and 1,089,321 short indels, with SVs covering 4.8 times more base pairs than SNPs. All three variant types revealed a highly congruent population structure and similar patterns of F ST and density variation along the genome. Finally, we performed outlier detection and redundancy analysis (RDA) to identify variants of interest in the putative local adaptation of Romaine and Puyjalon salmon. Genes located near these variants were enriched for biological processes related to nervous system function, suggesting that observed variation in traits such as age at smoltification could arise from differences in neural development. This study therefore demonstrates the feasibility of large-scale SV characterization and highlights its relevance for salmonid population genomics.

A survey of algorithms for the detection of genomic structural variants from long-read sequencing data

As long-read sequencing technologies are becoming increasingly popular, a number of methods have been developed for the discovery and analysis of structural variants (SVs) from long reads. Long reads enable detection of SVs that could not be previously detected from short-read sequencing, but computational methods must adapt to the unique challenges and opportunities presented by long-read sequencing. Here, we summarize over 50 long-read-based methods for SV detection, genotyping and visualization, and discuss how new telomere-to-telomere genome assemblies and pangenome efforts can improve the accuracy and drive the development of SV callers in the future.

REViewer: haplotype-resolved visualization of read alignments in and around tandem repeats

BACKGROUND

Expansions of short tandem repeats are the cause of many neurogenetic disorders including familial amyotrophic lateral sclerosis, Huntington disease, and many others. Multiple methods have been recently developed that can identify repeat expansions in whole genome or exome sequencing data. Despite the widely recognized need for visual assessment of variant calls in clinical settings, current computational tools lack the ability to produce such visualizations for repeat expansions. Expanded repeats are difficult to visualize because they correspond to large insertions relative to the reference genome and involve many misaligning and ambiguously aligning reads.

RESULTS

We implemented REViewer, a computational method for visualization of sequencing data in genomic regions containing long repeat expansions and FlipBook, a companion image viewer designed for manual curation of large collections of REViewer images. To generate a read pileup, REViewer reconstructs local haplotype sequences and distributes reads to these haplotypes in a way that is most consistent with the fragment lengths and evenness of read coverage. To create appropriate training materials for onboarding new users, we performed a concordance study involving 12 scientists involved in short tandem repeat research. We used the results of this study to create a user guide that describes the basic principles of using REViewer as well as a guide to the typical features of read pileups that correspond to low confidence repeat genotype calls. Additionally, we demonstrated that REViewer can be used to annotate clinically relevant repeat interruptions by comparing visual assessment results of 44 FMR1 repeat alleles with the results of triplet repeat primed PCR. For 38 of these alleles, the results of visual assessment were consistent with triplet repeat primed PCR.

CONCLUSIONS

Read pileup plots generated by REViewer offer an intuitive way to visualize sequencing data in regions containing long repeat expansions. Laboratories can use REViewer and FlipBook to assess the quality of repeat genotype calls as well as to visually detect interruptions or other imperfections in the repeat sequence and the surrounding flanking regions. REViewer and FlipBook are available under open-source licenses at https://github.com/illumina/REViewer and https://github.com/broadinstitute/flipbook respectively.

NPSV: A simulation-driven approach to genotyping structural variants in whole-genome sequencing data

BACKGROUND

Structural variants (SVs) play a causal role in numerous diseases but are difficult to detect and accurately genotype (determine zygosity) in whole-genome next-generation sequencing data. SV genotypers that assume that the aligned sequencing data uniformly reflect the underlying SV or use existing SV call sets as training data can only partially account for variant and sample-specific biases.

RESULTS

We introduce NPSV, a machine learning-based approach for genotyping previously discovered SVs that uses next-generation sequencing simulation to model the combined effects of the genomic region, sequencer, and alignment pipeline on the observed SV evidence. We evaluate NPSV alongside existing SV genotypers on multiple benchmark call sets. We show that NPSV consistently achieves or exceeds state-of-the-art genotyping accuracy across SV call sets, samples, and variant types. NPSV can specifically identify putative de novo SVs in a trio context and is robust to offset SV breakpoints.

CONCLUSIONS

Growing SV databases and the increasing availability of SV calls from long-read sequencing make stand-alone genotyping of previously identified SVs an increasingly important component of genome analyses. By treating potential biases as a "black box" that can be simulated, NPSV provides a framework for accurately genotyping a broad range of SVs in both targeted and genome-scale applications.

Samplot: a platform for structural variant visual validation and automated filtering

Visual validation is an important step to minimize false-positive predictions from structural variant (SV) detection. We present Samplot, a tool for creating images that display the read depth and sequence alignments necessary to adjudicate purported SVs across samples and sequencing technologies. These images can be rapidly reviewed to curate large SV call sets. Samplot is applicable to many biological problems such as SV prioritization in disease studies, analysis of inherited variation, or de novo SV review. Samplot includes a machine learning package that dramatically decreases the number of false positives without human review. Samplot is available at https://github.com/ryanlayer/samplot .

Ribbon: intuitive visualization for complex genomic variation

SUMMARY

Ribbon is an alignment visualization tool that shows how alignments are positioned within both the reference and read contexts, giving an intuitive view that enables a better understanding of structural variants and the read evidence supporting them. Ribbon was born out of a need to curate complex structural variant calls and determine whether each was well supported by long-read evidence, and it uses the same intuitive visualization method to shed light on contig alignments from genome-to-genome comparisons.

AVAILABILITY AND IMPLEMENTATION

Ribbon is freely available online at http://genomeribbon.com/ and is open-source at https://github.com/marianattestad/ribbon.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

ClinSV: clinical grade structural and copy number variant detection from whole genome sequencing data

Whole genome sequencing (WGS) has the potential to outperform clinical microarrays for the detection of structural variants (SV) including copy number variants (CNVs), but has been challenged by high false positive rates. Here we present ClinSV, a WGS based SV integration, annotation, prioritization, and visualization framework, which identified 99.8% of simulated pathogenic ClinVar CNVs > 10 kb and 11/11 pathogenic variants from matched microarrays. The false positive rate was low (1.5-4.5%) and reproducibility high (95-99%). In clinical practice, ClinSV identified reportable variants in 22 of 485 patients (4.7%) of which 35-63% were not detectable by current clinical microarray designs. ClinSV is available at https://github.com/KCCG/ClinSV .

Aquila enables reference-assisted diploid personal genome assembly and comprehensive variant detection based on linked reads

We introduce Aquila, a new approach to variant discovery in personal genomes, which is critical for uncovering the genetic contributions to health and disease. Aquila uses a reference sequence and linked-read data to generate a high quality diploid genome assembly, from which it then comprehensively detects and phases personal genetic variation. The contigs of the assemblies from our libraries cover >95% of the human reference genome, with over 98% of that in a diploid state. Thus, the assemblies support detection and accurate genotyping of the most prevalent types of human genetic variation, including single nucleotide polymorphisms (SNPs), small insertions and deletions (small indels), and structural variants (SVs), in all but the most difficult regions. All heterozygous variants are phased in blocks that can approach arm-level length. The final output of Aquila is a diploid and phased personal genome sequence, and a phased Variant Call Format (VCF) file that also contains homozygous and a few unphased heterozygous variants. Aquila represents a cost-effective approach that can be applied to cohorts for variation discovery or association studies, or to single individuals with rare phenotypes that could be caused by SVs or compound heterozygosity.

Unraveling the chaotic genomic landscape of primary and metastatic canine appendicular osteosarcoma with current sequencing technologies and bioinformatic approaches

Osteosarcoma is a rare disease in children but is one of the most common cancers in adult large breed dogs. The mutational landscape of both the primary and pulmonary metastatic tumor in two dogs with appendicular osteosarcoma (OSA) was comprehensively evaluated using an automated whole genome sequencing, exome and RNA-seq pipeline that was adapted for this study for use in dogs. Chromosomal lesions were the most common type of mutation. The mutational landscape varied substantially between dogs but the lesions within the same patient were similar. Copy number neutral loss of heterozygosity in mutant TP53 was the most significant driver mutation and involved a large region in the middle of chromosome 5. Canine and human OSA is characterized by loss of cell cycle checkpoint integrity and DNA damage response pathways. Mutational profiling of individual patients with canine OSA would be recommended prior to targeted therapy, given the heterogeneity seen in our study and previous studies.

Cyrius: accurate CYP2D6 genotyping using whole-genome sequencing data

Responsible for the metabolism of ~21% of clinically used drugs, CYP2D6 is a critical component of personalized medicine initiatives. Genotyping CYP2D6 is challenging due to sequence similarity with its pseudogene paralog CYP2D7 and a high number and variety of common structural variants (SVs). Here we describe a novel bioinformatics method, Cyrius, that accurately genotypes CYP2D6 using whole-genome sequencing (WGS) data. We show that Cyrius has superior performance (96.5% concordance with truth genotypes) compared to existing methods (84–86.8%). After implementing the improvements identified from the comparison against the truth data, Cyrius’s accuracy has since been improved to 99.3%. Using Cyrius, we built a haplotype frequency database from 2504 ethnically diverse samples and estimate that SV-containing star alleles are more frequent than previously reported. Cyrius will be an important tool to incorporate pharmacogenomics in WGS-based precision medicine initiatives.

Parliament2: Accurate structural variant calling at scale

BACKGROUND

Structural variants (SVs) are critical contributors to genetic diversity and genomic disease. To predict the phenotypic impact of SVs, there is a need for better estimates of both the occurrence and frequency of SVs, preferably from large, ethnically diverse cohorts. Thus, the current standard approach requires the use of short paired-end reads, which remain challenging to detect, especially at the scale of hundreds to thousands of samples.

FINDINGS

We present Parliament2, a consensus SV framework that leverages multiple best-in-class methods to identify high-quality SVs from short-read DNA sequence data at scale. Parliament2 incorporates pre-installed SV callers that are optimized for efficient execution in parallel to reduce the overall runtime and costs. We demonstrate the accuracy of Parliament2 when applied to data from NovaSeq and HiSeq X platforms with the Genome in a Bottle (GIAB) SV call set across all size classes. The reported quality score per SV is calibrated across different SV types and size classes. Parliament2 has the highest F1 score (74.27%) measured across the independent gold standard from GIAB. We illustrate the compute performance by processing all 1000 Genomes samples (2,691 samples) in <1 day on GRCH38. Parliament2 improves the runtime performance of individual methods and is open source (https://github.com/slzarate/parliament2), and a Docker image, as well as a WDL implementation, is available.

CONCLUSION

Parliament2 provides both a highly accurate single-sample SV call set from short-read DNA sequence data and enables cost-efficient application over cloud or cluster environments, processing thousands of samples.

A crowdsourced set of curated structural variants for the human genome

A high quality benchmark for small variants encompassing 88 to 90% of the reference genome has been developed for seven Genome in a Bottle (GIAB) reference samples. However a reliable benchmark for large indels and structural variants (SVs) is more challenging. In this study, we manually curated 1235 SVs, which can ultimately be used to evaluate SV callers or train machine learning models. We developed a crowdsourcing app-SVCurator-to help GIAB curators manually review large indels and SVs within the human genome, and report their genotype and size accuracy. SVCurator displays images from short, long, and linked read sequencing data from the GIAB Ashkenazi Jewish Trio son [NIST RM 8391/HG002]. We asked curators to assign labels describing SV type (deletion or insertion), size accuracy, and genotype for 1235 putative insertions and deletions sampled from different size bins between 20 and 892,149 bp. 'Expert' curators were 93% concordant with each other, and 37 of the 61 curators had at least 78% concordance with a set of 'expert' curators. The curators were least concordant for complex SVs and SVs that had inaccurate breakpoints or size predictions. After filtering events with low concordance among curators, we produced high confidence labels for 935 events. The SVCurator crowdsourced labels were 94.5% concordant with the heuristic-based draft benchmark SV callset from GIAB. We found that curators can successfully evaluate putative SVs when given evidence from multiple sequencing technologies.

Genome sequencing in cytogenetics: Comparison of short-read and linked-read approaches for germline structural variant detection and characterization

BACKGROUND

Structural variants (SVs) include copy number variants (CNVs) and apparently balanced chromosomal rearrangements (ABCRs). Genome sequencing (GS) enables SV detection at base-pair resolution, but the use of short-read sequencing is limited by repetitive sequences, and long-read approaches are not yet validated for diagnosis. Recently, 10X Genomics proposed Chromium, a technology providing linked-reads to reconstruct long DNA fragments and which could represent a good alternative. No study has compared short-read to linked-read technologies to detect SVs in a constitutional diagnostic setting yet. The aim of this work was to determine whether the 10X Genomics technology enables better detection and comprehension of SVs than short-read WGS.

METHODS

We included 13 patients carrying various SVs. Whole genome analyses were performed using paired-end HiSeq X sequencing with (linked-read strategy) or without (short-read strategy) Chromium library preparation. Two different bioinformatic pipelines were used: Variants are called using BreakDancer for short-read strategy and LongRanger for long-read strategy. Variant interpretations were first blinded.

RESULTS

The short-read strategy allowed diagnosis of known SV in 10/13 patients. After unblinding, the linked-read strategy identified 10/13 SVs, including one (patient 7) missed by the short-read strategy.

CONCLUSION

In conclusion, regarding the results of this study, 10X Genomics solution did not improve the detection and characterization of SV.

Visualization tools for human structural variations identified by whole-genome sequencing

Visualizing structural variations (SVs) is a critical step for finding associations between SVs and human traits or diseases. Given that there are many sequencing platforms used for SV identification and given that how best to visualize SVs together with other data, such as read alignments and annotations, depends on research goals, there are dozens of SV visualization tools designed for different research goals and sequencing platforms. Here, we provide a comprehensive survey of over 30 SV visualization tools to help users choose which tools to use. This review targets users who wish to visualize a set of SVs identified from the massively parallel sequencing reads of an individual human genome. We first categorize the ways in which SV visualization tools display SVs into ten major categories, which we denote as view modules. View modules allow readers to understand the features of each SV visualization tool quickly. Next, we introduce the features of individual SV visualization tools from several aspects, including whether SV views are integrated with annotations, whether long-read alignment is displayed, whether underlying data structures are graph-based, the type of SVs shown, whether auditing is possible, whether bird's eye view is available, sequencing platforms, and the number of samples. We hope that this review will serve as a guide for readers on the currently available SV visualization tools and lead to the development of new SV visualization tools in the near future.

De novo diploid genome assembly for genome-wide structural variant detection

Detection of structural variants (SVs) on the basis of read alignment to a reference genome remains a difficult problem. De novo assembly, traditionally used to generate reference genomes, offers an alternative for SV detection. However, it has not been applied broadly to human genomes because of fundamental limitations of short-fragment approaches and high cost of long-read technologies. We here show that 10× linked-read sequencing supports accurate SV detection. We examined variants in six de novo 10× assemblies with diverse experimental parameters from two commonly used human cell lines: NA12878 and NA24385. The assemblies are effective for detecting mid-size SVs, which were discovered by simple pairwise alignment of the assemblies’ contigs to the reference (hg38). Our study also shows that the base-pair level SV breakpoint accuracy is high, with a majority of SVs having precisely correct sizes and breakpoints. Setting the ancestral state of SV loci by comparing to ape orthologs allows inference of the actual molecular mechanism (insertion or deletion) causing the mutation. In about half of cases, the mechanism is the opposite of the reference-based call. We uncover 214 SVs that may have been maintained as polymorphisms in the human lineage since before our divergence from chimp. Overall, we show that de novo assembly of 10× linked-read data can achieve cost-effective SV detection for personal genomes.

MoMI-G: modular multi-scale integrated genome graph browser

Background

Genome graph is an emerging approach for representing structural variants on genomes with branches. For example, representing structural variants of cancer genomes as a genome graph is more natural than representing such genomes as differences from the linear reference genome. While more and more structural variants are being identified by long-read sequencing, many of them are difficult to visualize using existing structural variants visualization tools. To this end, visualization method for large genome graphs such as human cancer genome graphs is demanded.

Results

We developed MOdular Multi-scale Integrated Genome graph browser, MoMI-G, a web-based genome graph browser that can visualize genome graphs with structural variants and supporting evidences such as read alignments, read depth, and annotations. This browser allows more intuitive recognition of large, nested, and potentially more complex structural variations. MoMI-G has view modules for different scales, which allow users to view the whole genome down to nucleotide-level alignments of long reads. Alignments spanning reference alleles and those spanning alternative alleles are shown in the same view. Users can customize the view, if they are not satisfied with the preset views. In addition, MoMI-G has Interval Card Deck, a feature for rapid manual inspection of hundreds of structural variants. Herein, we describe the utility of MoMI-G by using representative examples of large and nested structural variations found in two cell lines, LC-2/ad and CHM1.

Conclusions

Users can inspect complex and large structural variations found by long-read analysis in large genomes such as human genomes more smoothly and more intuitively. In addition, users can easily filter out false positives by manually inspecting hundreds of identified structural variants with supporting long-read alignments and annotations in a short time.

Software availability

MoMI-G is freely available at https://github.com/MoMI-G/MoMI-G under the MIT license.

Comparison of INDEL Calling Tools with Simulation Data and Real Short-Read Data

Insertions and deletions (INDELs) comprise a significant proportion of human genetic variation, and recent papers have revealed that many human diseases may be attributable to INDELs. With the development of next-generation sequencing (NGS) technology, many statistical/computational tools have been developed for calling INDELs. However, there are differences among those tools, and comparisons among them have been limited. In order to better understand these inter-tool differences, five popular and publicly available INDEL calling tools-GATK HaplotypeCaller, Platypus, VarScan2, Scalpel, and GotCloud-were evaluated using simulation data, 1000 Genomes Project data, and family-based sequencing data. The accuracy of INDEL calling by each tool was mainly evaluated by concordance rates. Family-based sequencing data, which consisted of 49 individuals from eight Korean families, were used to calculate Mendelian error rates. Our comparison results show that GATK HaplotypeCaller usually performs the best and that joint calling with Platypus can lead to additional improvements in accuracy. The result of this study provides important information regarding future directions for the variant detection and the algorithms development.

Single-cell cloning of human T-cell lines reveals clonal variation in cell death responses to chemotherapeutics

Genetic modification of human leukemic cell lines using CRISPR-Cas9 has become a staple of gene-function studies. Single-cell cloning of modified cells is frequently used to facilitate studies of gene function. Inherent in this approach is an assumption that the genetic drift, amplified in some cell lines by mutations in DNA replication and repair machinery, as well as non-genetic factors will not introduce significant levels of experimental cellular heterogeneity in clones derived from parental populations. In this study, we characterize the variation in cell death of fifty clonal cell lines generated from human Jurkat and MOLT-4 T-cells edited by CRISPR-Cas9. We demonstrate a wide distribution of sensitivity to chemotherapeutics between non-edited clonal human leukemia T-cell lines, and also following CRISPR-Cas9 editing at the NLRP1 locus, or following transfection with non-targeting sgRNA controls. The cell death sensitivity profile of clonal cell lines was consistent across experiments and failed to revert to the non-clonal parental phenotype. Whole genome sequencing of two clonal cell lines edited by CRISPR-Cas9 revealed unique and shared genetic variants, which had minimal read support in the non-clonal parental population and were not suspected CRISPR-Cas9 off-target effects. These variants included genes related to cell death and drug metabolism. The variation in cell death phenotype of clonal populations of human T-cell lines may be a consequence of T-cell line genetic instability, and to a lesser extent clonal heterogeneity in the parental population or CRISPR-Cas9 off-target effects not predicted by current models. This work highlights the importance of genetic variation between clonal T-cell lines in the design, conduct, and analysis of experiments to investigate gene function after single-cell cloning.

SV-plaudit: A cloud-based framework for manually curating thousands of structural variants

SV-plaudit is a framework for rapidly curating structural variant (SV) predictions. For each SV, we generate an image that visualizes the coverage and alignment signals from a set of samples. Images are uploaded to our cloud framework where users assess the quality of each image using a client-side web application. Reports can then be generated as a tab-delimited file or annotated Variant Call Format (VCF) file. As a proof of principle, nine researchers collaborated for 1 hour to evaluate 1,350 SVs each. We anticipate that SV-plaudit will become a standard step in variant calling pipelines and the crowd-sourced curation of other biological results.Code available at https://github.com/jbelyeu/SV-plauditDemonstration video available at https://www.youtube.com/watch?v=ono8kHMKxDs.

Copy-number variants in clinical genome sequencing: deployment and interpretation for rare and undiagnosed disease

Purpose

Current diagnostic testing for genetic disorders involves serial use of specialized assays spanning multiple technologies. In principle, genome sequencing (GS) can detect all genomic pathogenic variant types on a single platform. Here we evaluate copy-number variant (CNV) calling as part of a clinically accredited GS test.

Methods

We performed analytical validation of CNV calling on 17 reference samples, compared the sensitivity of GS-based variants with those from a clinical microarray, and set a bound on precision using orthogonal technologies. We developed a protocol for family-based analysis of GS-based CNV calls, and deployed this across a clinical cohort of 79 rare and undiagnosed cases.

Results

We found that CNV calls from GS are at least as sensitive as those from microarrays, while only creating a modest increase in the number of variants interpreted (~10 CNVs per case). We identified clinically significant CNVs in 15% of the first 79 cases analyzed, all of which were confirmed by an orthogonal approach. The pipeline also enabled discovery of a uniparental disomy (UPD) and a 50% mosaic trisomy 14. Directed analysis of select CNVs enabled breakpoint level resolution of genomic rearrangements and phasing of de novo CNVs.

Conclusion

Robust identification of CNVs by GS is possible within a clinical testing environment.

Prioritisation of structural variant calls in cancer genomes

Sensitivity of short read DNA-sequencing for gene fusion detection is improving, but is hampered by the significant amount of noise composed of uninteresting or false positive hits in the data. In this paper we describe a tiered prioritisation approach to extract high impact gene fusion events from existing structural variant calls. Using cell line and patient DNA sequence data we improve the annotation and interpretation of structural variant calls to best highlight likely cancer driving fusions. We also considerably improve on the automated visualisation of the high impact structural variants to highlight the effects of the variants on the resulting transcripts. The resulting framework greatly improves on readily detecting clinically actionable structural variants.

Clinical detection of deletion structural variants in whole-genome sequences

Optimal management of acutely ill infants with monogenetic diseases requires rapid identification of causative haplotypes. Whole-genome sequencing (WGS) has been shown to identify pathogenic nucleotide variants in such infants. Deletion structural variants (DSVs, >50 nt) are implicated in many genetic diseases, and tools have been designed to identify DSVs using short-read WGS. Optimisation and integration of these tools into a WGS pipeline could improve diagnostic sensitivity and specificity of WGS. In addition, it may improve turnaround time when compared with current CNV assays, enhancing utility in acute settings. Here we describe DSV detection methods for use in WGS for rapid diagnosis in acutely ill infants: SKALD (Screening Konsensus and Annotation of Large Deletions) combines calls from two tools (Breakdancer and GenomeStrip) with calibrated filters and clinical interpretation rules. In four WGS runs, the average analytic precision (positive predictive value) of SKALD was 78%, and recall (sensitivity) was 27%, when compared with validated reference DSV calls. When retrospectively applied to a cohort of 36 families with acutely ill infants SKALD identified causative DSVs in two. The first was heterozygous deletion of exons 1–3 of MMP21 in trans with a heterozygous frame-shift deletion in two siblings with transposition of the great arteries and heterotaxy. In a newborn female with dysmorphic features, ventricular septal defect and persistent pulmonary hypertension, SKALD identified the breakpoints of a heterozygous, de novo 1p36.32p36.13 deletion. In summary, consensus DSV calling, implemented in an 8-h computational pipeline with parameterised filtering, has the potential to increase the diagnostic yield of WGS in acutely ill neonates and discover novel disease genes.

svclassify: a method to establish benchmark structural variant calls

Background

The human genome contains variants ranging in size from small single nucleotide polymorphisms (SNPs) to large structural variants (SVs). High-quality benchmark small variant calls for the pilot National Institute of Standards and Technology (NIST) Reference Material (NA12878) have been developed by the Genome in a Bottle Consortium, but no similar high-quality benchmark SV calls exist for this genome. Since SV callers output highly discordant results, we developed methods to combine multiple forms of evidence from multiple sequencing technologies to classify candidate SVs into likely true or false positives. Our method (svclassify) calculates annotations from one or more aligned bam files from many high-throughput sequencing technologies, and then builds a one-class model using these annotations to classify candidate SVs as likely true or false positives.

Results

We first used pedigree analysis to develop a set of high-confidence breakpoint-resolved large deletions. We then used svclassify to cluster and classify these deletions as well as a set of high-confidence deletions from the 1000 Genomes Project and a set of breakpoint-resolved complex insertions from Spiral Genetics. We find that likely SVs cluster separately from likely non-SVs based on our annotations, and that the SVs cluster into different types of deletions. We then developed a supervised one-class classification method that uses a training set of random non-SV regions to determine whether candidate SVs have abnormal annotations different from most of the genome. To test this classification method, we use our pedigree-based breakpoint-resolved SVs, SVs validated by the 1000 Genomes Project, and assembly-based breakpoint-resolved insertions, along with semi-automated visualization using svviz.

Conclusions

We find that candidate SVs with high scores from multiple technologies have high concordance with PCR validation and an orthogonal consensus method MetaSV (99.7 % concordant), and candidate SVs with low scores are questionable. We distribute a set of 2676 high-confidence deletions and 68 high-confidence insertions with high svclassify scores from these call sets for benchmarking SV callers. We expect these methods to be particularly useful for establishing high-confidence SV calls for benchmark samples that have been characterized by multiple technologies.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-016-2366-2) contains supplementary material, which is available to authorized users.