Long-read nanopore sequencing enables accurate confirmation of a recurrent PMS2 insertion-deletion variant located in a region of complex genomic architecture.

Targeted next generation sequencing (NGS) is the predominant methodology for the molecular genetic diagnosis of inherited conditions. In many laboratories, NGS-identified variants are routinely validated using a different method, to minimize the risk of a false-positive diagnosis. This can be particularly important when pathogenic variants are located in complex genomic regions. In this situation, new long-read sequencing technologies have potential advantages over existing alternatives. However, practical examples of their utility for diagnostic purposes remain scant. Here, we report the use of nanopore sequencing to validate a PMS2 mutation refractory to conventional methods. In a patient who presented with colorectal cancer and loss of PMS2 immunostaining, short-read NGS of Lynch syndrome-associated genes identified the recurrent PMS2 insertion-deletion variant, c.736_741delinsTGTGTGTGAAG (p.Pro246Cysfs*3). Confirmation of this variant using bidirectional Sanger sequencing was impeded by an upstream intron 6 poly(T) tract. Using a locus-specific amplicon template, we undertook nanopore long-read sequencing in order to assess the diagnostic accuracy of this platform. Pairwise comparison between a curated benchmark allele (derived from short-read NGS and unidirectional Sanger sequencing) and the consensus nanopore dataset revealed 100% sequence identity. Our experience provides insight into the robustness and ease of deployment of "third-generation" sequencing for accurate characterisation of pathogenic variants.

Improving the identification of patients with a genetic diagnosis of familial hypercholesterolaemia in primary care: A strategy to achieve the NHS long term plan.

BACKGROUND AND AIMS: We aimed to validate a nurse-led process using electronic health records to identify those at risk of familial hypercholesterolaemia (FH) for genetic diagnosis in primary care. METHODS: Those at risk of FH were identified using searches developed and refined locally and implemented in primary care by a trained nurse; they were invited for further assessment and genetic testing if indicated. Family members at risk of FH were identified and invited for cascade testing. RESULTS: In total 94,444 patient records were screened (expected prevalence of FH (1 in 250); 377). Of 176 records which already had a diagnostic for FH, 15 had been genetically confirmed and one was undergoing DNA testing. A further 572 (0.61%) were identified as high risk of FH. After desktop screening, 113 (15%) were invited for further assessment. Of these, 73 individuals attended the primary care clinic (64%) of whom 61 (54%) underwent proband genetic testing. Pathogenic variants were detected in 22 cases (36%) and variants of unknown significance in a further 4 cases; a total of 26 probands (43%) were therefore referred for family cascade testing. CONCLUSIONS: An optimised FH identification pathway, based on the NICE CG71 recommendations for systematic searching of primary care electronic health records, can be deployed successfully in primary care settings.

Assessing the utility of long-read nanopore sequencing for rapid and efficient characterization of mobile element insertions.

Short-read next generation sequencing (NGS) has become the predominant first-line technique used to diagnose patients with rare genetic conditions. Inherent limitations of short-read technology, notably for the detection and characterization of complex insertion-containing variants, are offset by the ability to concurrently screen many disease genes. "Third-generation" long-read sequencers are increasingly being deployed as an orthogonal adjunct technology, but their full potential for molecular genetic diagnosis has yet to be exploited. Here, we describe three diagnostic cases in which pathogenic mobile element insertions were refractory to characterization by short-read sequencing. To validate the accuracy of the long-read technology, we first used Sanger sequencing to confirm the integration sites and derive curated benchmark sequences of the variant-containing alleles. Long-read nanopore sequencing was then performed on locus-specific amplicons. Pairwise comparison between these data and the previously determined benchmark alleles revealed 100% identity of the variant-containing sequences. We demonstrate a number of technical advantages over existing wet-laboratory approaches, including in silico size selection of a mixed pool of amplification products, and the relative ease with which an automated informatics workflow can be established. Our findings add to a growing body of literature describing the diagnostic utility of long-read sequencing.

Long-read nanopore sequencing resolves a TMEM231 gene conversion event causing Meckel-Gruber syndrome.

The diagnostic deployment of massively parallel short-read next-generation sequencing (NGS) has greatly improved genetic test availability, speed, and diagnostic yield, particularly for rare inherited disorders. Nonetheless, diagnostic approaches based on short-read sequencing have a poor ability to accurately detect gene conversion events. We report on the genetic analysis of a family in which 3 fetuses had clinical features consistent with the autosomal recessive disorder Meckel-Gruber syndrome (MKS). Targeted NGS of 29 known MKS-associated genes revealed a heterozygous TMEM231 splice donor variant c.929+1A>G. Comparative read-depth analysis, performed to identify a second pathogenic allele, revealed an apparent heterozygous deletion of TMEM231 exon 4. To verify this result we performed single-molecule long-read sequencing of a long-range polymerase chain reaction product spanning this locus. We identified four missense variants that were absent from the short-read dataset due to the preferential mapping of variant-containing reads to a downstream TMEM231 pseudogene. Consistent with the parental segregation analysis, we demonstrate that the single-molecule long reads could be used to show that the variants are arranged in trans. Our experience shows that robust validation of apparent dosage variants remains essential to avoid the pitfalls of short-read sequencing and that new third-generation long-read sequencing technologies can already aid routine clinical care.

Whole Exon Deletion in the GFAP Gene Is a Novel Molecular Mechanism Causing Alexander Disease.

Alexander disease (AD) is a leukodystrophy caused by heterozygous mutations in the gene encoding the glial fibrillary acidic protein (GFAP). Currently, de novo heterozygous missense mutations in the GFAP gene are identified in over 95% of patients with AD. However, patients with biopsy-proven AD have been reported in whom no GFAP mutation has been identified. We report identical twin boys presenting in infancy with seizures and developmental delay in whom MR appearances were suggestive of AD with the exception of an unusual, bilateral, arc of calcification at the frontal white-gray junction. Initial mutation screening of the GFAP gene did not identify a mutation. Whole exome sequencing in both brothers revealed a de novo heterozygous in-frame deletion of the whole of exon 5 of the GFAP gene. Mutations in the GFAP gene are thought to result in a toxic effect of mutant GFAP disrupting the formation of the normal intermediate filament network and resulting in Rosenthal fiber formation, which has hitherto not been linked to exonic scale copy number variants in GFAP. Further studies on mutation negative AD patients are warranted to determine whether a similar mechanism underlies their disease.

Increased Sensitivity of Diagnostic Mutation Detection by Re-analysis Incorporating Local Reassembly of Sequence Reads.

BACKGROUND: Diagnostic genetic testing programmes based on next-generation DNA sequencing have resulted in the accrual of large datasets of targeted raw sequence data. Most diagnostic laboratories process these data through an automated variant-calling pipeline. Validation of the chosen analytical methods typically depends on confirming the detection of known sequence variants. Despite improvements in short-read alignment methods, current pipelines are known to be comparatively poor at detecting large insertion/deletion mutations. METHODS: We performed clinical validation of a local reassembly tool, ABRA (assembly-based realigner), through retrospective reanalysis of a cohort of more than 2000 hereditary cancer cases. RESULTS: ABRA enabled detection of a 96-bp deletion, 4-bp insertion mutation in PMS2 that had been initially identified using a comparative read-depth approach. We applied an updated pipeline incorporating ABRA to the entire cohort of 2000 cases and identified one previously undetected pathogenic variant, a 23-bp duplication in PTEN. We demonstrate the effect of read length on the ability to detect insertion/deletion variants by comparing HiSeq2500 (2 × 101-bp) and NextSeq500 (2 × 151-bp) sequence data for a range of variants and thereby show that the limitations of shorter read lengths can be mitigated using appropriate informatics tools. CONCLUSIONS: This work highlights the need for ongoing development of diagnostic pipelines to maximize test sensitivity. We also draw attention to the large differences in computational infrastructure required to perform day-to-day versus large-scale reprocessing tasks.

Characterization and Genomic Localization of a SMAD4 Processed Pseudogene.

Like many clinical diagnostic laboratories, the Yorkshire Regional Genetics Service undertakes routine investigation of cancer-predisposed individuals by high-throughput sequencing of patient DNA that has been target-enriched for genes associated with hereditary cancer. Accurate diagnosis using such reagents requires alertness regarding rare nonpathogenic variants that may interfere with variant calling. In a cohort of 2042 such cases, we identified 5 that initially appeared to be carriers of a 95-bp deletion of SMAD4 intron 6. More detailed analysis indicated that these individuals all carried one copy of a SMAD4 processed gene. Because of its interference with diagnostic analysis, we characterized this processed gene in detail. Whole-genome sequencing and confirmatory Sanger sequencing of junction PCR products were used to show that in each of the 5 cases, the SMAD4 processed gene was integrated at the same position on chromosome 9, located within the last intron of the SCAI gene. This rare polymorphic processed gene therefore reflects the occurrence of a single ancestral retrotransposition event. Compared to the reference SMAD4 mRNA sequence NM_005359.5 (https://www.ncbi.nlm.nih.gov/nucleotide), the 5' and 3' untranslated regions of the processed gene are both truncated, but its open reading frame is unaltered. Our experience leads us to advocate the use of an RNA-seq aligner as part of diagnostic assay quality assurance, since this allows recognition of processed pseudogenes in a comparatively facile automated fashion.

Robust diagnostic genetic testing using solution capture enrichment and a novel variant-filtering interface.

Targeted hybridization enrichment prior to next-generation sequencing is a widespread method for characterizing sequence variation in a research setting, and is being adopted by diagnostic laboratories. However, the number of variants identified can overwhelm clinical laboratories with strict time constraints, the final interpretation of likely pathogenicity being a particular bottleneck. To address this, we have developed an approach in which, after automatic variant calling on a standard unix pipeline, subsequent variant filtering is performed interactively, using AgileExomeFilter and AgilePindelFilter (http://dna.leeds.ac.uk/agile), tools designed for clinical scientists with standard desktop computers. To demonstrate the method's diagnostic efficacy, we tested 128 patients using (1) a targeted capture of 36 cancer-predisposing genes or (2) whole-exome capture for diagnosis of the genetically heterogeneous disorder primary ciliary dyskinesia (PCD). In the cancer cohort, complete concordance with previous diagnostic data was achieved across 793 variant genotypes. A high yield (42%) was also achieved for exome-based PCD diagnosis, underscoring the scalability of our method. Simple adjustments to the variant filtering parameters further allowed the identification of a homozygous truncating mutation in a presumptive new PCD gene, DNAH8. These tools should allow diagnostic laboratories to expand their testing portfolios flexibly, using a standard set of reagents and techniques.

GeneScreen: a program for high-throughput mutation detection in DNA sequence electropherograms.

BACKGROUND: While massively parallel DNA sequencing methods continue to evolve rapidly, the benchmark technique for detection and verification of rare (particularly disease-causing) sequence variants remains four-colour dye-terminator sequencing by capillary electrophoresis. The high throughput and long read lengths currently available have shifted the bottleneck in mutation detection away from data generation to data analysis. While excellent computational methods have been developed for quantifying sequence accuracy and detecting variants, either during de novo sequence assembly or for single-nucleotide polymorphism detection, the identification, verification and annotation of very rare sequence variants remains a rather labour-intensive process for which few software aids exist. AIM: To provide a freely available, intuitive software application for highly efficient mutation screening of large sequence batches. METHODS AND RESULTS: The authors developed GeneScreen, a desktop program that analyses capillary electropherograms and compares their sequences with a known reference for identification of mutations. The detected sequence variants are then made available for rapid assessment and annotation via a graphical user interface, allowing chosen variants to be exported for reporting and archiving. The program was validated using more than 16,000 diagnostic laboratory sequence traces. CONCLUSION: Using GeneScreen, a single user requires only a few minutes to identify rare mutations in hundreds of sequence traces, with comparable sensitivity to expensive commercial products.

Genetic diagnosis of familial breast cancer using clonal sequencing.

Using conventional Sanger sequencing as a reference standard, we compared the sensitivity, specificity, and capacity of the Illumina GA II platform for the detection of TP53, BRCA1, and BRCA2 mutations in established tumor cell lines and DNA from patients with germline mutations. A total of 656 coding variants were identified in four cell lines and 65 patient DNAs. All of the known pathogenic mutations (including point mutations and insertions/deletions of up to 16 nucleotides) were identified, using a combination of the Illumina data analysis pipeline with custom and commercial sequence alignment software. In our configuration, clonal sequencing outperforms current diagnostic methods, providing a reduction in analysis times and in reagent costs compared with conventional sequencing. These improvements open the possibility of BRCA1/2 testing for a wider spectrum of at-risk women, and will allow the genetic classification of tumors prior to the use of novel PARP inhibitors to treat BRCA-deficient breast cancers.

A high utility integrated map of the pig genome.

BACKGROUND: The domestic pig is being increasingly exploited as a system for modeling human disease. It also has substantial economic importance for meat-based protein production. Physical clone maps have underpinned large-scale genomic sequencing and enabled focused cloning efforts for many genomes. Comparative genetic maps indicate that there is more structural similarity between pig and human than, for example, mouse and human, and we have used this close relationship between human and pig as a way of facilitating map construction. RESULTS: Here we report the construction of the most highly continuous bacterial artificial chromosome (BAC) map of any mammalian genome, for the pig (Sus scrofa domestica) genome. The map provides a template for the generation and assembly of high-quality anchored sequence across the genome. The physical map integrates previous landmark maps with restriction fingerprints and BAC end sequences from over 260,000 BACs derived from 4 BAC libraries and takes advantage of alignments to the human genome to improve the continuity and local ordering of the clone contigs. We estimate that over 98% of the euchromatin of the 18 pig autosomes and the X chromosome along with localized coverage on Y is represented in 172 contigs, with chromosome 13 (218 Mb) represented by a single contig. The map is accessible through pre-Ensembl, where links to marker and sequence data can be found. CONCLUSION: The map will enable immediate electronic positional cloning of genes, benefiting the pig research community and further facilitating use of the pig as an alternative animal model for human disease. The clone map and BAC end sequence data can also help to support the assembly of maps and genome sequences of other artiodactyls.