Search

Tandem repeat (TR) variation is associated with gene expression changes and over 50 rare monogenic diseases. Recent advances in sequencing have enabled accurate, long reads that can characterize the full-length sequence and methylation profile of TRs. However, despite these advances in sequencing technology, computational methods to fully profile tandem repeats across the genome do not exist. To address this gap, we introduce tools for tandem repeat genotyping (TRGT), visualization and an accompanying TR database. TRGT accurately resolves the length and sequence composition of TR regions in the human genome. Assessing 937,122 TRs, TRGT showed a Mendelian concordance of 99.56%, allowing a single repeat unit difference. In six samples with known repeat expansions, TRGT detected all repeat expansions while also identifying methylation signals, mosaicism, and providing finer resolution of repeat length. Additionally, we release a database with allele sequences and methylation levels for 937,122 TRs across 100 genomes.

In studies of families with rare disease, it is common to screen for de novo mutations, as well as recessive or dominant variants that explain the phenotype. However, the filtering strategies and software used to prioritize high-confidence variants vary from study to study. In an effort to establish recommendations for rare disease research, we explore effective guidelines for variant (SNP and INDEL) filtering and report the expected number of candidates for de novo dominant, recessive, and autosomal dominant modes of inheritance. We derived these guidelines using two large family-based cohorts that underwent whole-genome sequencing, as well as two family cohorts with whole-exome sequencing. The filters are applied to common attributes, including genotype-quality, sequencing depth, allele balance, and population allele frequency. The resulting guidelines yield ~10 candidate SNP and INDEL variants per exome, and 18 per genome for recessive and de novo dominant modes of inheritance, with substantially more candidates for autosomal dominant inheritance. For family-based, whole-genome sequencing studies, this number includes an average of three de novo, ten compound heterozygous, one autosomal recessive, four X-linked variants, and roughly 100 candidate variants following autosomal dominant inheritance. The slivar software we developed to establish and rapidly apply these filters to VCF files is available at https://github.com/brentp/slivar under an MIT license, and includes documentation and recommendations for best practices for rare disease analysis.

Expansions of short tandem repeats (STRs) cause dozens of rare Mendelian diseases. However, STR expansions, especially those arising from repeats not present in the reference genome, are challenging to detect from short-read sequencing data. Such “novel” STRs include new repeat units occurring at known STR loci, or entirely new STR loci where the sequence is absent from the reference genome. A primary cause of difficulty detecting STR expansions is that reads arising from STR expansions are frequently mismapped or unmapped. To address this challenge, we have developed STRling, a new STR detection algorithm that counts k-mers (short DNA sequences of length k) in DNA sequencing reads, to efficiently recover reads that inform the presence and size of STR expansions. As a result, STRling can call expansions at both known and novel STR loci. STRling has a sensitivity of 83% for 14 known STR disease loci, including the novel STRs that cause CANVAS and DBQD2. It is the first method to resolve the position of novel STR expansions to base pair accuracy. Such accuracy is essential to interpreting the consequence of each expansion. STRling has an estimated 0.078 false discovery rate for known pathogenic loci in unaffected individuals and a 0.20 false discovery rate for genome-wide loci in unaffected individuals when using variants called from long-read data as truth. STRling is fast, scalable on cloud computing, open-source, and freely available at https://github.com/quinlan-lab/STRling.

In studies of families with rare disease, it is common to screen for de novo mutations, as well as recessive or dominant variants that explain the phenotype. However, the filtering strategies and software used to prioritize high-confidence variants vary from study to study. In an effort to establish recommendations for rare disease research, we derive effective guidelines for variant filtering and report the expected number of candidates for de novo dominant and recessive modes of inheritance. The filters are applied to common attributes, including genotype quality, sequencing depth, allele balance, and population allele frequency. The resulting guidelines yield approximately 10 candidate SNP and INDEL variants per exome, and 19 per genome. For whole genomes, this includes an average of three de novo, ten compound-heterozygotes, one autosomal recessive, four X-linked variants, and roughly 100 candidate variants following autosomal dominant inheritance. The slivar software we developed to establish and rapidly apply these filters to VCF files is available at https://github.com/brentp/slivar under an MIT license, and includes documentation and recommendations for best practices for rare disease analysis.

Rapid identification of clinically significant variants is key to the successful application of next generation sequencing technologies in clinical practice. The Melbourne Genomics Health Alliance (MGHA) variant prioritization framework employs a gene prioritization index based on clinician-generated a priori gene lists, and a variant prioritization index (VPI) based on rarity, conservation and protein effect. We used data from 80 patients who underwent singleton whole exome sequencing (WES) to test the ability of the framework to rank causative variants highly, and compared it against the performance of other gene and variant prioritization tools. Causative variants were identified in 59 of the patients. Using the MGHA prioritization framework the average rank of the causative variant was 2.24, with 76% ranked as the top priority variant, and 90% ranked within the top five. Using clinician-generated gene lists resulted in ranking causative variants an average of 8.2 positions higher than prioritization based on variant properties alone. This clinically driven prioritization approach significantly outperformed purely computational tools, placing a greater proportion of causative variants top or in the top 5 (permutation P-value=0.001). Clinicians included 40 of the 49 WES diagnoses in their a priori list of differential diagnoses (81%). The lists generated by PhenoTips and Phenomizer contained 14 (29%) and 18 (37%) of these diagnoses respectively. These results highlight the benefits of clinically led variant prioritization in increasing the efficiency of singleton WES data analysis and have important implications for developing models for the funding and delivery of genomic services.

Background Bioinformatics software tools are often created ad hoc, frequently by people without extensive training in software development. In particular, for beginners, the barrier to entry in bioinformatics software development is high, especially if they want to adopt good programming practices. Even experienced developers do not always follow best practices. This results in the proliferation of poorer-quality bioinformatics software, leading to limited scalability and inefficient use of resources; lack of reproducibility, usability, adaptability, and interoperability; and erroneous or inaccurate results. Findings We have developed Bionitio, a tool that automates the process of starting new bioinformatics software projects following recommended best practices. With a single command, the user can create a new well-structured project in 1 of 12 programming languages. The resulting software is functional, carrying out a prototypical bioinformatics task, and thus serves as both a working example and a template for building new tools. Key features include command-line argument parsing, error handling, progress logging, defined exit status values, a test suite, a version number, standardized building and packaging, user documentation, code documentation, a standard open source software license, software revision control, and containerization. Conclusions Bionitio serves as a learning aid for beginner-to-intermediate bioinformatics programmers and provides an excellent starting point for new projects. This helps developers adopt good programming practices from the beginning of a project and encourages high-quality tools to be developed more rapidly. This also benefits users because tools are more easily installed and consistent in their usage. Bionitio is released as open source software under the MIT License and is available at https://github.com/bionitio-team/bionitio.

In the clinical setting, exome sequencing has become standard-of-care in diagnosing rare genetic disorders, however many patients remain unsolved. Trio sequencing has been demonstrated to produce a higher diagnostic yield than singleton (proband-only) sequencing. Parental sequencing is especially useful when a disease is suspected to be caused by a de novo variant in the proband, because parental data provide a strong filter for the majority of variants that are shared by the proband and their parents. However the additional cost of sequencing the parents makes the trio strategy uneconomical for many clinical situations. With two thirds of the sequencing budget being spent on parents, these are funds that could be used to sequence more probands. For this reason many clinics are reluctant to sequence parents.Here we propose a pooled-parent strategy for exome sequencing of individuals with likely de novo disease. In this strategy, DNA from all the parents of a cohort of unrelated probands is pooled together into a single exome capture and sequencing run. Variants called in the proband can then be filtered if they are also found in the parent pool, resulting in a shorter list of prioritised variants. To evaluate the pooled-parent strategy we performed a series of simulations by combining reads from individual exomes to imitate sample pooling. We assessed the recall and false positive rate and investigated the trade-off between pool size and recall rate. We compared the performance of GATK HaplotypeCaller individual and joint calling, and FreeBayes to genotype pooled samples. Finally, we applied a pooled-parent strategy to a set of real unsolved cases and showed that the parent pool is a powerful filter that is complementary to other commonly used variant filters such as population variant frequencies.

Short tandem repeat (STR) expansions have been identified as the causal DNA mutation in dozens of Mendelian diseases. Historically, pathogenic STR expansions could only be detected by single locus techniques, such as PCR and electrophoresis. The ability to use short read sequencing data to screen for STR expansions has the potential to reduce both the time and cost to reaching diagnosis and enable the discovery of new causal STR loci. Most existing tools detect STR variation within the read length, and so are unable to detect the majority of pathogenic expansions. Those tools that can detect large expansions are limited to a set of known disease loci and as yet no new disease causing STR expansions have been identified with high-throughput sequencing technologies.Here we address this by presenting STRetch, a new genome-wide method to detect STR expansions at all loci across the human genome. We demonstrate the use of STRetch for detecting pathogenic STR expansions in short-read whole genome sequencing data with a very low false discovery rate. We further demonstrate the application of STRetch to solve cases of patients with undiagnosed disease and apply STRetch to the analysis of 97 whole genomes to reveal variation at STR loci. STRetch assesses expansions at all STR loci in the genome and allows screening for novel disease-causing STRs.STRetch is open source software, available fromgithub.com/Oshlack/STRetch.

The benefits of implementing high throughput sequencing in the clinic are quickly becoming apparent. However, few freely available bioinformatics pipelines have been built from the ground up with clinical genomics in mind. Here we present Cpipe, a pipeline designed specifically for clinical genetic disease diagnostics. Cpipe was developed by the Melbourne Genomics Health Alliance, an Australian initiative to promote common approaches to genomics across healthcare institutions. As such, Cpipe has been designed to provide fast, effective and reproducible analysis, while also being highly flexible and customisable to meet the individual needs of diverse clinical settings. Cpipe is being shared with the clinical sequencing community as an open source project and is available at http://cpipeline.org. Electronic supplementary material The online version of this article (doi:10.1186/s13073-015-0191-x) contains supplementary material, which is available to authorized users.

'Big data' analytics can be defined by the requirement for flexible, high throughput computational analysis methods applied to large, heterogeneous datasets. We propose an architectural approach to 'big data' challenges in which the movement of data is minimized, and analysis methods are implemented on the data as portable services. We term this approach 'data interlocking'. We demonstrate the feasibility of this approach through a domain specific implementation of a data interlocking architecture, in which an on-demand computational workbench provides portable high-throughput analysis methods to large genomic datasets on cloud infrastructure.

Rapid molecular typing of bacterial pathogens is critical for public health epidemiology, surveillance and infection control, yet routine use of whole genome sequencing (WGS) for these purposes poses significant challenges. Here we present SRST2, a read mapping-based tool for fast and accurate detection of genes, alleles and multi-locus sequence types (MLST) from WGS data. Using >900 genomes from common pathogens, we show SRST2 is highly accurate and outperforms assembly-based methods in terms of both gene detection and allele assignment. We include validation of SRST2 within a public health laboratory, and demonstrate its use for microbial genome surveillance in the hospital setting. In the face of rising threats of antimicrobial resistance and emerging virulence among bacterial pathogens, SRST2 represents a powerful tool for rapidly extracting clinically useful information from raw WGS data. Source code is available from http://katholt.github.io/srst2/. Electronic supplementary material The online version of this article (doi:10.1186/s13073-014-0090-6) contains supplementary material, which is available to authorized users.