Search

BACKGROUND: Tandem repeats comprise significant proportion of the human genome including coding and regulatory regions. They are highly prone to repeat number variation and nucleotide mutation due to their repetitive and unstable nature, making them a major source of genomic variation between individuals. Despite recent advances in high throughput sequencing, analysis of tandem repeats in the context of complex diseases is still hindered by technical limitations. We report a novel targeted sequencing approach, which allows simultaneous analysis of hundreds of repeats. We developed a Bayesian algorithm, namely - GtTR - which combines information from a reference long-read dataset with a short read counting approach to genotype tandem repeats at population scale. PCR sizing analysis was used for validation. RESULTS: We used a PacBio long-read sequenced sample to generate a reference tandem repeat genotype dataset with on average 13% absolute deviation from PCR sizing results. Using this reference dataset GtTR generated estimates of VNTR copy number with accuracy within 95% high posterior density (HPD) intervals of 68 and 83% for capture sequence data and 200X WGS data respectively, improving to 87 and 94% with use of a PCR reference. We show that the genotype resolution increases as a function of depth, such that the median 95% HPD interval lies within 25, 14, 12 and 8% of the its midpoint copy number value for 30X, 200X WGS, 395X and 800X capture sequence data respectively. We validated nine targets by PCR sizing analysis and genotype estimates from sequencing results correlated well with PCR results. CONCLUSIONS: The novel genotyping approach described here presents a new cost-effective method to explore previously unrecognized class of repeat variation in GWAS studies of complex diseases at the population level. Further improvements in accuracy can be obtained by improving accuracy of the reference dataset.

BACKGROUND: Detection of genomic inversions remains challenging. Many existing methods primarily target inzversions with a non repetitive breakpoint, leaving inverted repeat (IR) mediated non-allelic homologous recombination (NAHR) inversions largely unexplored. RESULT: We present npInv, a novel tool specifically for detecting and genotyping NAHR inversion using long read sub-alignment of long read sequencing data. We benchmark npInv with other tools in both simulation and real data. We use npInv to generate a whole-genome inversion map for NA12878 consisting of 30 NAHR inversions (of which 15 are novel), including all previously known NAHR mediated inversions in NA12878 with flanking IR less than 7kb. Our genotyping accuracy on this dataset was 94%. We used PCR to confirm the presence of two of these novel inversions. We show that there is a near linear relationship between the length of flanking IR and the minimum inversion size, without inverted repeats. CONCLUSION: The application of npInv shows high accuracy in both simulation and real data. The results give deeper insight into understanding inversion.

MOTIVATION: Targeted sequencing using capture probes has become increasingly popular in clinical applications due to its scalability and cost-effectiveness. The approach also allows for higher sequencing coverage of the targeted regions resulting in better analysis statistical power. However, because of the dynamics of the hybridization process, it is difficult to evaluate the efficiency of the probe design prior to the experiments which are time consuming and costly. RESULTS: We developed CapSim, a software package for simulation of targeted sequencing. Given a genome sequence and a set of probes, CapSim simulates the fragmentation, the dynamics of probe hybridization and the sequencing of the captured fragments on Illumina and PacBio sequencing platforms. The simulated data can be used for evaluating the performance of the analysis pipeline, as well as the efficiency of the probe design. Parameters of the various stages in the sequencing process can also be evaluated in order to optimize the experiments. AVAILABILITY AND IMPLEMENTATION: CapSim is publicly available under BSD license at https://github.com/Devika1/capsim. CONTACT: l.coin@imb.uq.edu.au. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Extensively drug-resistant Klebsiella pneumoniae (XDR-KP) infections cause high mortality and are disseminating globally. Identifying the genetic basis underpinning resistance allows for rapid diagnosis and treatment. XDR isolates sourced from Greece and Brazil, including 19 polymyxin-resistant and five polymyxin-susceptible strains, were subjected to whole genome sequencing. Seventeen of the 19 polymyxin-resistant isolates harboured variations upstream or within mgrB. The most common mutation identified was an insertion at nucleotide position 75 in mgrB via an ISKpn26-like element in the ST258 lineage and ISKpn13 in one ST11 isolate. Three strains acquired an IS1 element upstream of mgrB and another strain had an ISKpn25 insertion at 133 bp. Other isolates had truncations (C28STOP, Q30STOP) or a missense mutation (D29E) affecting mgrB. Complementation assays revealed all mgrB perturbations contributed to resistance. Missense mutations in phoQ (T281M, G385C) were also found to facilitate resistance. Several variants in phoPQ co-segregating with the ISKpn26-like insertion were identified as potential partial suppressor mutations. Three ST258 samples were found to contain subpopulations with different resistance-conferring mutations, including the ISKpn26-like insertion colonizing with a novel mutation in pmrB (P158R), both confirmed via complementation assays. These findings highlight the broad spectrum of chromosomal modifications which can facilitate and regulate resistance against polymyxins in K. pneumoniae.

The majority of human chromosome ends remain incompletely assembled due to their highly repetitive structure. In this study, we use BioNano data to anchor and extend chromosome ends from two European trios as well as two unrelated Asian genomes. At least 11 BioNano assembled chromosome ends are structurally divergent from the reference genome, including both missing sequence and extensions. These extensions are heritable and in some cases divergent between Asian and European samples. Six out of nine predicted extension sequences from NA12878 can be confirmed and filled by nanopore data. We identify two multi-kilobase sequence families both enriched more than 100-fold in extension sequence (p-values < 1e-5) whose origins can be traced to interstitial sequence on ancestral primate chromosome 7. Extensive sub-telomeric duplication of these families has occurred in the human lineage subsequent to divergence from chimpanzees.

Sequencing by translocating DNA fragments through an array of nanopores is a rapidly maturing technology that offers faster and cheaper sequencing than other approaches. However, accurately deciphering the DNA sequence from the noisy and complex electrical signal is challenging. Here, we report Chiron, the first deep learning model to achieve end-to-end basecalling and directly translate the raw signal to DNA sequence without the error-prone segmentation step. Trained with only a small set of 4,000 reads, we show that our model provides state-of-the-art basecalling accuracy, even on previously unseen species. Chiron achieves basecalling speeds of more than 2,000 bases per second using desktop computer graphics processing units.

Third generation sequencing technologies provide the opportunity to improve genome assemblies by generating long reads spanning most repeat sequences. However, current analysis methods require substantial amounts of sequence data and computational resources to overcome the high error rates. Furthermore, they can only perform analysis after sequencing has completed, resulting in either over-sequencing, or in a low quality assembly due to under-sequencing. Here we present npScarf, which can scaffold and complete short read assemblies while the long read sequencing run is in progress. It reports assembly metrics in real-time so the sequencing run can be terminated once an assembly of sufficient quality is obtained. In assembling four bacterial and one eukaryotic genomes, we show that npScarf can construct more complete and accurate assemblies while requiring less sequencing data and computational resources than existing methods. Our approach offers a time- and resource-effective strategy for completing short read assemblies.

The recently introduced Oxford Nanopore MinION platform generates DNA sequence data in real-time. This has great potential to shorten the sample-to-results time and is likely to have benefits such as rapid diagnosis of bacterial infection and identification of drug resistance. However, there are few tools available for streaming analysis of real-time sequencing data. Here, we present a framework for streaming analysis of MinION real-time sequence data, together with probabilistic streaming algorithms for species typing, strain typing and antibiotic resistance profile identification. Using four culture isolate samples, as well as a mixed-species sample, we demonstrate that bacterial species and strain information can be obtained within 30 min of sequencing and using about 500 reads, initial drug-resistance profiles within two hours, and complete resistance profiles within 10 h. While strain identification with multi-locus sequence typing required more than 15x coverage to generate confident assignments, our novel gene-presence typing could detect the presence of a known strain with 0.5x coverage. We also show that our pipeline can process over 100 times more data than the current throughput of the MinION on a desktop computer.