Comprehensive de novo mutation discovery with HiFi long-read sequencing

BACKGROUND

Long-read sequencing (LRS) techniques have been very successful in identifying structural variants (SVs). However, the high error rate of LRS made the detection of small variants (substitutions and short indels < 20 bp) more challenging. The introduction of PacBio HiFi sequencing makes LRS also suited for detecting small variation. Here we evaluate the ability of HiFi reads to detect de novo mutations (DNMs) of all types, which are technically challenging variant types and a major cause of sporadic, severe, early-onset disease.

METHODS

We sequenced the genomes of eight parent-child trios using high coverage PacBio HiFi LRS (~ 30-fold coverage) and Illumina short-read sequencing (SRS) (~ 50-fold coverage). De novo substitutions, small indels, short tandem repeats (STRs) and SVs were called in both datasets and compared to each other to assess the accuracy of HiFi LRS. In addition, we determined the parent-of-origin of the small DNMs using phasing.

RESULTS

We identified a total of 672 and 859 de novo substitutions/indels, 28 and 126 de novo STRs, and 24 and 1 de novo SVs in LRS and SRS respectively. For the small variants, there was a 92 and 85% concordance between the platforms. For the STRs and SVs, the concordance was 3.6 and 0.8%, and 4 and 100% respectively. We successfully validated 27/54 LRS-unique small variants, of which 11 (41%) were confirmed as true de novo events. For the SRS-unique small variants, we validated 42/133 DNMs and 8 (19%) were confirmed as true de novo event. Validation of 18 LRS-unique de novo STR calls confirmed none of the repeat expansions as true DNM. Confirmation of the 23 LRS-unique SVs was possible for 19 candidate SVs of which 10 (52.6%) were true de novo events. Furthermore, we were able to assign 96% of DNMs to their parental allele with LRS data, as opposed to just 20% with SRS data.

CONCLUSIONS

HiFi LRS can now produce the most comprehensive variant dataset obtainable by a single technology in a single laboratory, allowing accurate calling of substitutions, indels, STRs and SVs. The accuracy even allows sensitive calling of DNMs on all variant levels, and also allows for phasing, which helps to distinguish true positive from false positive DNMs.

Long-Read Sequencing Identifies Novel Pathogenic Intronic Variants in Gitelman Syndrome

BACKGROUND

Gitelman syndrome is a salt-losing tubulopathy characterized by hypokalemic alkalosis and hypomagnesemia. It is caused by homozygous recessive or compound heterozygous pathogenic variants in SLC12A3 , which encodes the Na + -Cl - cotransporter (NCC). In up to 10% of patients with Gitelman syndrome, current genetic techniques detect only one specific pathogenic variant. This study aimed to identify a second pathogenic variant in introns, splice sites, or promoters to increase the diagnostic yield.

METHODS

Long-read sequencing of SLC12A3 was performed in 67 DNA samples from individuals with suspected Gitelman syndrome in whom a single likely pathogenic or pathogenic variant was previously detected. In addition, we sequenced DNA samples from 28 individuals with one variant of uncertain significance or no candidate variant. Midigene splice assays assessed the pathogenicity of novel intronic variants.

RESULTS

A second likely pathogenic/pathogenic variant was identified in 45 (67%) patients. Those with two likely pathogenic/pathogenic variants had a more severe electrolyte phenotype than other patients. Of the 45 patients, 16 had intronic variants outside of canonic splice sites (nine variants, mostly deep intronic, six novel), whereas 29 patients had an exonic variant or canonic splice site variant. Midigene splice assays of the previously known c.1670-191C>T variant and intronic candidate variants demonstrated aberrant splicing patterns.

CONCLUSION

Intronic pathogenic variants explain an important part of the missing heritability in Gitelman syndrome. Long-read sequencing should be considered in diagnostic workflows for Gitelman syndrome.

DeNovoCNN: a deep learning approach to de novo variant calling in next generation sequencing data

De novo mutations (DNMs) are an important cause of genetic disorders. The accurate identification of DNMs from sequencing data is therefore fundamental to rare disease research and diagnostics. Unfortunately, identifying reliable DNMs remains a major challenge due to sequence errors, uneven coverage, and mapping artifacts. Here, we developed a deep convolutional neural network (CNN) DNM caller (DeNovoCNN), that encodes the alignment of sequence reads for a trio as 160\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$ \times$\end{document}164 resolution images. DeNovoCNN was trained on DNMs of 5616 whole exome sequencing (WES) trios achieving total 96.74% recall and 96.55% precision on the test dataset. We find that DeNovoCNN has increased recall/sensitivity and precision compared to existing DNM calling approaches (GATK, DeNovoGear, DeepTrio, Samtools) based on the Genome in a Bottle reference dataset and independent WES and WGS trios. Validations of DNMs based on Sanger and PacBio HiFi sequencing confirm that DeNovoCNN outperforms existing methods. Most importantly, our results suggest that DeNovoCNN is likely robust against different exome sequencing and analyses approaches, thereby allowing the application on other datasets. DeNovoCNN is freely available as a Docker container and can be run on existing alignment (BAM/CRAM) and variant calling (VCF) files from WES and WGS without a need for variant recalling.

Optical genome mapping identifies a germline retrotransposon insertion in SMARCB1 in two siblings with atypical teratoid rhabdoid tumors

In a subset of pediatric cancers, a germline cancer predisposition is highly suspected based on clinical and pathological findings, but genetic evidence is lacking, which hampers genetic counseling and predictive testing in the families involved. We describe a family with two siblings born from healthy parents who were both neonatally diagnosed with atypical teratoid rhabdoid tumor (ATRT). This rare and aggressive pediatric tumor is associated with biallelic inactivation of SMARCB1, and in 30% of the cases, a predisposing germline mutation is involved. Whereas the tumors of both siblings showed loss of expression of SMARCB1 and acquired homozygosity of the locus, whole exome and whole genome sequencing failed to identify germline or somatic SMARCB1 pathogenic mutations. We therefore hypothesized that the insertion of a pathogenic repeat‐rich structure might hamper its detection, and we performed optical genome mapping (OGM) as an alternative strategy to identify structural variation in this locus. Using this approach, an insertion of ~2.8 kb within intron 2 of SMARCB1 was detected. Long‐range PCR covering this region remained unsuccessful, but PacBio HiFi genome sequencing identified this insertion to be a SINE‐VNTR‐Alu, subfamily E (SVA‐E) retrotransposon element, which was present in a mosaic state in the mother. This SVA‐E insertion disrupts correct splicing of the gene, resulting in loss of a functional allele. This case demonstrates the power of OGM and long‐read sequencing to identify genomic variations in high‐risk cancer‐predisposing genes that are refractory to detection with standard techniques, thereby completing the clinical and molecular diagnosis of such complex cases and greatly improving counseling and surveillance of the families involved. © 2021 The Authors. The Journal of Pathology published by John Wiley & Sons, Ltd. on behalf of The Pathological Society of Great Britain and Ireland.

Long-read trio sequencing of individuals with unsolved intellectual disability

Long-read sequencing (LRS) has the potential to comprehensively identify all medically relevant genome variation, including variation commonly missed by short-read sequencing (SRS) approaches. To determine this potential, we performed LRS around 15×-40× genome coverage using the Pacific Biosciences Sequel I System for five trios. The respective probands were diagnosed with intellectual disability (ID) whose etiology remained unresolved after SRS exomes and genomes. Systematic assessment of LRS coverage showed that ~35 Mb of the human reference genome was only accessible by LRS and not SRS. Genome-wide structural variant (SV) calling yielded on average 28,292 SV calls per individual, totaling 12.9 Mb of sequence. Trio-based analyses which allowed to study segregation, showed concordance for up to 95% of these SV calls across the genome, and 80% of the LRS SV calls were not identified by SRS. De novo mutation analysis did not identify any de novo SVs, confirming that these are rare events. Because of high sequence coverage, we were also able to call single nucleotide substitutions. On average, we identified 3 million substitutions per genome, with a Mendelian inheritance concordance of up to 97%. Of these, ~100,000 were located in the ~35 Mb of the genome that was only captured by LRS. Moreover, these variants affected the coding sequence of 64 genes, including 32 known Mendelian disease genes. Our data show the potential added value of LRS compared to SRS for identifying medically relevant genome variation.

De Novo Variants in CNOT1, a Central Component of the CCR4-NOT Complex Involved in Gene Expression and RNA and Protein Stability, Cause Neurodevelopmental Delay

CNOT1 is a member of the CCR4-NOT complex, which is a master regulator, orchestrating gene expression, RNA deadenylation, and protein ubiquitination. We report on 39 individuals with heterozygous de novo CNOT1 variants, including missense, splice site, and nonsense variants, who present with a clinical spectrum of intellectual disability, motor delay, speech delay, seizures, hypotonia, and behavioral problems. To link CNOT1 dysfunction to the neurodevelopmental phenotype observed, we generated variant-specific Drosophila models, which showed learning and memory defects upon CNOT1 knockdown. Introduction of human wild-type CNOT1 was able to rescue this phenotype, whereas mutants could not or only partially, supporting our hypothesis that CNOT1 impairment results in neurodevelopmental delay. Furthermore, the genetic interaction with autism-spectrum genes, such as ASH1L, DYRK1A, MED13, and SHANK3, was impaired in our Drosophila models. Molecular characterization of CNOT1 variants revealed normal CNOT1 expression levels, with both mutant and wild-type alleles expressed at similar levels. Analysis of protein-protein interactions with other members indicated that the CCR4-NOT complex remained intact. An integrated omics approach of patient-derived genomics and transcriptomics data suggested only minimal effects on endonucleolytic nonsense-mediated mRNA decay components, suggesting that de novo CNOT1 variants are likely haploinsufficient hypomorph or neomorph, rather than dominant negative. In summary, we provide strong evidence that de novo CNOT1 variants cause neurodevelopmental delay with a wide range of additional co-morbidities. Whereas the underlying pathophysiological mechanism warrants further analysis, our data demonstrate an essential and central role of the CCR4-NOT complex in human brain development.

BRCA Testing by Single-Molecule Molecular Inversion Probes

BACKGROUND

Despite advances in next generation DNA sequencing (NGS), NGS-based single gene tests for diagnostic purposes require improvements in terms of completeness, quality, speed, and cost. Single-molecule molecular inversion probes (smMIPs) are a technology with unrealized potential in the area of clinical genetic testing. In this proof-of-concept study, we selected 2 frequently requested gene tests, those for the breast cancer genes BRCA1 and BRCA2, and developed an automated work flow based on smMIPs.

METHODS

The BRCA1 and BRCA2 smMIPs were validated using 166 human genomic DNA samples with known variant status. A generic automated work flow was built to perform smMIP-based enrichment and sequencing for BRCA1, BRCA2, and the checkpoint kinase 2 (CHEK2) c.1100del variant.

RESULTS

Pathogenic and benign variants were analyzed in a subset of 152 previously BRCA-genotyped samples, yielding an analytical sensitivity and specificity of 100%. Following automation, blind analysis of 65 in-house samples and 267 Norwegian samples correctly identified all true-positive variants (>3000), with no false positives. Consequent to process optimization, turnaround times were reduced by 60% to currently 10-15 days. Copy number variants were detected with an analytical sensitivity of 100% and an analytical specificity of 88%.

CONCLUSIONS

smMIP-based genetic testing enables automated and reliable analysis of the coding sequences of BRCA1 and BRCA2. The use of single-molecule tags, double-tiled targeted enrichment, and capturing and sequencing in duplo, in combination with automated library preparation and data analysis, results in a robust process and reduces routine turnaround times. Furthermore, smMIP-based copy number variation analysis could make independent copy number variation tools like multiplex ligation-dependent probes amplification dispensable.

Targeted next-generation sequencing of a 12.5 Mb homozygous region reveals ANO10 mutations in patients with autosomal-recessive cerebellar ataxia

Autosomal-recessive cerebellar ataxias comprise a clinically and genetically heterogeneous group of neurodegenerative disorders. In contrast to their dominant counterparts, unraveling the molecular background of these ataxias has proven to be more complicated and the currently known mutations provide incomplete coverage for genotyping of patients. By combining SNP array-based linkage analysis and targeted resequencing of relevant sequences in the linkage interval with the use of next-generation sequencing technology, we identified a mutation in a gene and have shown its association with autosomal-recessive cerebellar ataxia. In a Dutch consanguineous family with three affected siblings a homozygous 12.5 Mb region on chromosome 3 was targeted by array-based sequence capture. Prioritization of all detected sequence variants led to four candidate genes, one of which contained a variant with a high base pair conservation score (phyloP score: 5.26). This variant was a leucine-to-arginine substitution in the DUF 590 domain of a 16K transmembrane protein, a putative calcium-activated chloride channel encoded by anoctamin 10 (ANO10). The analysis of ANO10 by Sanger sequencing revealed three additional mutations: a homozygous mutation (c.1150_1151del [p.Leu384fs]) in a Serbian family and a compound-heterozygous splice-site mutation (c.1476+1G>T) and a frameshift mutation (c.1604del [p.Leu535X]) in a French family. This illustrates the power of using initial homozygosity mapping with next-generation sequencing technology to identify genes involved in autosomal-recessive diseases. Moreover, identifying a putative calcium-dependent chloride channel involved in cerebellar ataxia adds another pathway to the list of pathophysiological mechanisms that may cause cerebellar ataxia.