ISMapper: identifying transposase insertion sites in bacterial genomes from short read sequence data

The challenge of detecting indels in bacterial genomes from short-read sequencing data

Structural variations (SVs) are differences in genomic regions that are larger than 1 kilobase-pair (Kbp) between individuals, and can arise from errant DNA repair mechanisms, whole genome duplications, and transposable element activity across the genome. Recent advances, optimizations, and cost reductions in next generation sequencing technologies have facilitated the exponential increase in the amount of available short read genomic data. Here we present SaVor, a flexible, reproducible SV calling workflow that accepts single or multi-lane short-read paired-end Illumina sequence data, or BAM files as input to generate a consensus SV call-set based on user-provided merge parameters. We tested SaVor on 1,165 Arabidopsis thaliana whole genome sequences and benchmarked its performance on a set of SVs derived from the same accessions using Lumpy. Intersection calls i.e. SVs supported by 3 SV callers showed the highest precision (>0.91) while union calls supported by at least 1 caller showed the highest recall (>0.88). We found that the former suffers from decreased recall (<0.51) and the latter decreased precision (<0.57). Depending on the merge strategy, trade-offs in recall and precision need to be considered for downstream analyses of SV call-sets from short-read data. SaVor is an open-source Snakemake pipeline and is available on GitHub at https://github.com/ChabbyTMD/SaVor

Elucidation of the genomic organizations of transgene insertion sites is essential for the genetic studies of transgenic plants. Herein, we establish an analysis pipeline that identifies the transgene insertion sites as well as the presence of vector backbones, through de novo genome assembly with high-throughput sequencing data in two transgenic soybean lines, AtYUCCA6-#5 and 35S-UGT72E3/2-#7. Sequencing data of approximately 28× and 29× genome coverages for each line generated by high-throughput sequencing were de novo assembled. The databases generated from the de novo assembled sequences were used to search contigs that contained putative insertion sites and their flanking sequences (integration sites) of transgene fragments using transgenic vector sequences as queries. The predicted integration site sequences, which are located at three annotated genes that might regulate plant development or confer disease resistance, were then confirmed by local alignment against the soybean reference genome and PCR amplification. As results, we revealed the precise transgene-flanking sequences and sequence rearrangements at insertion sites in both the transgenic lines, as well as the aberrant insertion of a transgene fragment. Consequently, relative to experimental or enrichment technologies, our approach is straightforward and time-effective, providing an alternative method for the identification of insertion sites in transgenic plants.

BackgroundBacterial whole-genome sequencing based on short-read technologies often results in a draft assembly formed by contiguous sequences. The introduction of long-read sequencing technologies permits those contiguous sequences to be unambiguously bridged into complete genomes. However, the elevated costs associated with long-read sequencing frequently limit the number of bacterial isolates that can be long-read sequenced. Here we evaluated the recently released 96 barcoding kit from Oxford Nanopore Technologies (ONT) to generate complete genomes on a high-throughput basis. In addition, we propose an isolate selection strategy that optimizes a representative selection of isolates for long-read sequencing considering as input large-scale bacterial collections.ResultsDespite an uneven distribution of long reads per barcode, near-complete chromosomal sequences (assembly contiguity = 0.89) were generated for 96 Escherichia coli isolates with associated short-read sequencing data. The assembly contiguity of the plasmid replicons was even higher (0.98), which indicated the suitability of the multiplexing strategy for studies focused on resolving plasmid sequences. We benchmarked hybrid and ONT-only assemblies and showed that the combination of ONT sequencing data with short-read sequencing data is still highly desirable (i) to perform an unbiased selection of isolates for long-read sequencing, (ii) to achieve an optimal genome accuracy and completeness, and (iii) to include small plasmids underrepresented in the ONT library.ConclusionsThe proposed long-read isolate selection ensures the completion of bacterial genomes that span the genome diversity inherent in large collections of bacterial isolates. We show the potential of using this multiplexing approach to close bacterial genomes on a high-throughput basis.

CpG methylation, a key epigenetic mark involved in gene regulation, development, and other biological processes, is commonly analyzed using Whole-Genome Bisulfite Sequencing (WGBS). However, bisulfite treatment causes significant DNA degradation. Enzymatic Methyl-seq (EM-seq) offers a short-read alternative that preserves DNA integrity but requires conversion steps, limiting its compatibility with downstream analyses. Third-generation sequencing technologies, such as Oxford Nanopore Technologies (ONT) and PacBio, enable direct detection of DNA modifications without altering the DNA, providing simultaneous genome and epigenome information. This work presents a comprehensive dataset combining long- and short-read sequencing data, including ONT, PacBio, Enzymatic Methyl-seq, and WGBS, for two agronomically relevant species: pig (Sus scrofa) and quail (Coturnix japonica). Data quality evaluation reveals high nucleotide quality scores for PacBio and short reads, robust alignment rates for long reads, and inter-method correlations in CpG methylation calling ranging from 0.76 to 0.99. This dataset is a valuable resource for training methylation callers and represents the first combined methylation dataset for these species, providing an essential benchmark for assessing emerging sequencing technologies.

Copy Number Alterations (CNAs) such as deletions and duplications; compose a larger percentage of genetic variations than single nucleotide polymorphisms or other structural variations in cancer genomes that undergo major chromosomal re-arrangements. It is, therefore, imperative to identify cancer-specific somatic copy number alterations (SCNAs), with respect to matched normal tissue, in order to understand their association with the disease. We have devised an accurate, sensitive, and easy-to-use tool, COPS, COpy number using Paired Samples, for detecting SCNAs. We rigorously tested the performance of COPS using short sequence simulated reads at various sizes and coverage of SCNAs, read depths, read lengths and also with real tumor:normal paired samples. We found COPS to perform better in comparison to other known SCNA detection tools for all evaluated parameters, namely, sensitivity (detection of true positives), specificity (detection of false positives) and size accuracy. COPS performed well for sequencing reads of all lengths when used with most upstream read alignment tools. Additionally, by incorporating a downstream boundary segmentation detection tool, the accuracy of SCNA boundaries was further improved. Here, we report an accurate, sensitive and easy to use tool in detecting cancer-specific SCNAs using short-read sequence data. In addition to cancer, COPS can be used for any disease as long as sequence reads from both disease and normal samples from the same individual are available. An added boundary segmentation detection module makes COPS detected SCNA boundaries more specific for the samples studied. COPS is available at ftp://115.119.160.213 with username “cops” and password “cops”.

Quantitative monitoring of emerging antimicrobial resistance genes (ARGs) using short-read sequences remains challenging due to the high frequency of amino acid functional domains and motifs shared with related but functionally distinct (non-target) proteins. To facilitate ARG monitoring efforts using unassembled short reads, we present novel ROCker models for mcr, mph, erm, and lnu ARG families, as well as models for variants of special public health concern within these families, including mcr-1, mphA, ermB, lnuF, lnuB, and lnuG genes. For this, we curated target gene sequence sets for model training and built these models using the recently updated ROCker V2 pipeline (Gerhardt et al., in review). To validate our models, we simulated reads from the whole genome of ARG-carrying isolates spanning a range of common read lengths and used them to challenge the filtering efficacy of ROCker versus common static filtering approaches, such as similarity searches using BLASTx with various e-value thresholds or hidden Markov models. ROCker models consistently showed F1 scores up to 10× higher (31% higher on average) and lower false-positive (by 30%, on average) and false-negative (by 16%, on average) rates based on 250 bp reads compared to alternative methods. The ROCker models and all related reference materials and data are freely available through http://enve-omics.ce.gatech.edu/rocker/models, further expanding the available model collection previously developed for other genes. Their application to short-read metagenomes, metatranscriptomes, and PCR amplicon data should facilitate more accurate classification and quantification of unassembled short-read sequences for these ARG families and specific genes.IMPORTANCEAntimicrobial resistance gene families encoding erm and mph genes confer resistance to the macrolide class of antimicrobials, which are used to treat a wide range of infections. Similarly, the mcr gene family confers resistance to polymyxin E (colistin), a drug of last resort for many serious drug-resistant bacterial infections, and the lnu gene family confers resistance to lincomycin, which is reserved for patients allergic to penicillin or where bacteria have developed resistance to other antimicrobials. Assessing the prevalence of these genes in clinical or environmental samples and monitoring their spread to new pathogens are thus important for quantifying the associated public health risk. However, detecting these and other resistance genes in short-read sequence data is technically challenging. Our ROCker bioinformatic pipeline achieves reliable detection and typing of broad-range target gene sequences in complex data sets, thus contributing toward solving an important problem in ongoing surveillance efforts of antimicrobial resistance.

The reconstruction of genomes using mapping based approaches with short reads experiences difficulties when resolving repetitive regions. These repetitive regions in genomes result in low mapping qualities of the respective reads, which in turn lead to many unresolved bases of the genotypers. Currently, the reconstruction of these regions is often based on modified references in which the repetitive regions are masked. However, for many references such masked genomes are not available or are based on repetitive regions of other genomes. Our idea is to identify repetitive regions in the reference genome de novo. These regions can then be used to reconstruct them separately using short read sequencing data. Afterwards the reconstructed repetitive sequence can be inserted into the reconstructed genome. We present the program DACCOR, which performs these steps automatically. Our results show an increased base pair resolution of the repetitive regions in the reconstruction of Treponema pallidum samples, resulting in fewer unresolved bases.

The reconstruction of genomes using mapping-based approaches with short reads experiences difficulties when resolving repetitive regions. These repetitive regions in genomes result in low mapping qualities of the respective reads, which in turn lead to many unresolved bases. Currently, the reconstruction of these regions is often based on modified references in which the repetitive regions are masked. However, for many references, such masked genomes are not available or are based on repetitive regions of other genomes. Our idea is to identify repetitive regions in the reference genome de novo. These regions can then be used to reconstruct them separately using short read sequencing data. Afterward, the reconstructed repetitive sequence can be inserted into the reconstructed genome. We present the program detection, characterization, and reconstruction of repetitive regions, which performs these steps automatically. Our results show an increased base pair resolution of the repetitive regions in the reconstruction of Treponema pallidum samples, resulting in fewer unresolved bases.

Differences in the prevalence of IS 6110 insertion sites in Mycobacterium tuberculosis strains: low and high copy number of IS 6110

Detection of KMT2A Partial Tandem Duplication in Myeloid Samples By Pacbio® Long Read Sequencing

Tandem duplication of coding sequence is an important mechanism of somatic mutation capable of activating oncogenes such as fms-related tyrosine kinase 3 (FLT3) as well as the potential for underlying constitutional or germline disease. Detection of internal tandem duplication (ITD) in short-read sequencing data remains challenging due to deficiencies in short-read alignment where reads carrying signatures of ITD are often unmapped. Assembly based approaches collapse repetitive sequences, resulting in reduced or lost genomic complexity. The majority of structural variation detection tools are limited to finding insertions that are contained entirely within individual read alignments, and are unable to detect read-spanning duplications. Tools designed to detect large structural variation from paired-end reads that map discordantly based on expected fragment size and orientation do not provide the precision needed to detect ITDs. Only tools that consider the characteristic alignment features that signal the presence of ITDs will be able to detect them; these alignment features include: an excess of reads with unaligned, soft-clipped bases, read pairs where one end is unmapped located near the site of insertion, and read pairs where both ends are unmapped. We introduce ITD Assembler, a novel approach that searches all unmapped reads from exome capture or whole genome sequencing data to identify duplications. From the entire set of unmapped reads plus subset of mapped reads harboring soft-clipped bases, de Bruijn graph assembly is applied to select reads that form cycles, indicative of duplicated sequence structures. Reads from de Bruijn graph cycles are then assembled using an Overlap Layout Consensus (OLC) methodology thereby alleviating the collapse of repeat sequences affecting de Bruijn graph assembly approaches. Resulting OLC assembled contigs are locally aligned to the reference sequence to annotate the position of ITDs. ITD Assembler was run on The Cancer Genome Atlas (TCGA) acute myeloid leukemia (AML) dataset, and FLT3-ITD detection rates were compared against orthogonal algorithms. ITD Assembler identified the highest percentage of TCGA FLT3-ITDs, reported significantly higher allele fractions, and discovered additional ITDs in the KIT, CEBPA and WT1 cancer genes. Citation Format: Oliver A. Hampton, Navin Rustagi, Jie Li, Liu Xi, Richard A. Gibbs, Sharon E. Plon, Marek Kimmel, David A. Wheeler. ITD Assembler: An algorithm for internal tandem duplication discovery from short-read sequencing data. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4856. doi:10.1158/1538-7445.AM2015-4856

As we enter the age of personalized medicine, healthcare is increasingly focused on tailoring diagnoses and treatments based on patients’ genetic and environmental circumstances. A critical component of a person's physiological makeup is their immune system, but individual genetic variation in many immune system genes has remained resistant to analysis using classical whole-genome or targeted sequencing approaches. In particular, germline adaptive immune system genes, like immunoglobulin (IG) and T cell receptor (TR) genes, are particularly hard to genotype using classic reference-based methods owing to their highly repetitive and homologous nature. In this paper, we present ImmunoTyper2, a new computational toolkit for genotyping the variable genes of the IG lambda and kappa, and the TR loci with short-read whole genome sequence data, using an integer linear programming formulation, as an update to the ImmunoTyper-SR suite, which focused on IGHV region only. We evaluate its genotyping performance using Mendelian concordance analysis in 590 trios from the 1000 Genomes Project, benchmarking 40 samples against HPRC assembly-derived genotypes, and assessing robustness through sequencing depth analysis and parameter sensitivity tests. We introduce allele call confidence metrics to help quantify reliability. We also perform a prospective disease association study, applying ImmunoTyper2 to a WGS data set from a cohort of 461 COVID-19 patients from the COVNET Consortium to demonstrate how it can be applied to investigate genetic associations with disease.

Obtaining complete and accurate bacterial genomes is vital for studying the characteristics of uncultured bacteria. Single-cell genomics is a promising approach for the culture-independent recovery of bacterial genomes from individual cells. However, single-amplified genomes (SAGs) often have fragmented and incomplete sequences due to chimeric and biased sequences introduced during the genome amplification process. To address this, we developed a single-cell amplified genome long-read assembly (scALA) workflow to construct complete circular SAGs (cSAGs) from long-read single-cell sequencing data of uncultured bacteria. We used the SAG-gel platform, which is both cost-effective and high-throughput, to obtain hundreds of short-read and long-read sequencing data for specific bacterial strains. The scALA workflow generated cSAGs by repeated in silico processing for sequence bias reduction and contig assembly. From 12 human fecal samples, including two cohabitant groups, scALA generated 16 cSAGs of three specifically targeted bacterial species: Anaerostipes hadrus, Agathobacter rectalis, and Ruminococcus gnavus. We discovered strain-specific structural variations shared among cohabiting hosts, while all cSAGs of the same species showed high homology in aligned genomic regions. A. hadrus cSAGs exhibited 10 kbp-long phage insertions, various saccharide metabolic capabilities, and different CRISPR-Cas systems in each strain. The sequence similarity of A. hadrus genomes did not necessarily correspond with orthologous functional genes, while host geographical regionality seemed to be highly related to gene possession. scALA allowed us to obtain closed circular genomes of specifically targeted bacteria from human microbiota samples, leading to an understanding of within-species diversities, including structural variations and linking mobile genetic elements, such as phages, to hosts. These analyses provide insight into microbial evolution, the adaptation of the community to environmental changes, and interactions with hosts. cSAGs constructed using this method can expand bacterial genome databases and our understanding of within-species diversities in uncultured bacteria.

Reference genomes of important cereals, including barley, emmer wheat and bread wheat, were released recently. Their comparison with genome size estimates obtained by flow cytometry indicated that the assemblies represent not more than 88–98% of the complete genome. This work is aimed at identifying the missing parts in two cereal genomes and proposing techniques to make the assemblies more complete. We focused on tandemly organised repetitive sequences, known to be underrepresented in genome assemblies generated from short-read sequence data. Our study found arrays of three tandem repeats with unit sizes of 1242 to 2726 bp present in the bread wheat reference genome generated from short reads. However, this and another wheat genome assembly employing long PacBio reads failed in integrating correctly the 2726-bp repeat in the pseudomolecule context. This suggests that tandem repeats of this size, frequently incorporated in unassigned scaffolds, may contribute to shrinking of pseudomolecules without reducing size of the entire assembly. We demonstrate how this missing information may be added to the pseudomolecules with the aid of nanopore sequencing of individual BAC clones and optical mapping. Using the latter technique, we identified and localised a 470-kb long array of 45S ribosomal DNA absent from the reference genome of barley.