Sequence deeper without sequencing more: Bayesian resolution of ambiguously mapped reads

Next-generation sequencing (NGS) has transformed molecular biology and contributed to many seminal insights into genomic regulation and function. Apart from whole-genome sequencing, an NGS workflow involves alignment of the sequencing reads to the genome of study, after which the resulting alignments can be used for downstream analyses. However, alignment is complicated by the repetitive sequences; many reads align to more than one genomic locus, with 15–30% of the genome not being uniquely mappable by short-read NGS. This problem is typically addressed by discarding reads that do not uniquely map to the genome, but this practice can lead to systematic distortion of the data. Previous studies that developed methods for handling ambiguously mapped reads were often of limited applicability or were computationally intensive, hindering their broader usage. In this work, we present SmartMap: an algorithm that augments industry-standard aligners to enable usage of ambiguously mapped reads by assigning weights to each alignment with Bayesian analysis of the read distribution and alignment quality. SmartMap is computationally efficient, utilizing far fewer weighting iterations than previously thought necessary to process alignments and, as such, analyzing more than a billion alignments of NGS reads in approximately one hour on a desktop PC. By applying SmartMap to peak-type NGS data, including MNase-seq, ChIP-seq, and ATAC-seq in three organisms, we can increase read depth by up to 53% and increase the mapped proportion of the genome by up to 18% compared to analyses utilizing only uniquely mapped reads. We further show that SmartMap enables the analysis of more than 140,000 repetitive elements that could not be analyzed by traditional ChIP-seq workflows, and we utilize this method to gain insight into the epigenetic regulation of different classes of repetitive elements. These data emphasize both the dangers of discarding ambiguously mapped reads and their power for driving biological discovery.     

Comparison of Hepatocellular Carcinoma miRNA Expression Profiling as Evaluated by Next Generation Sequencing and Microarray

MicroRNA (miRNA) expression profiling has proven useful in diagnosing and understanding the development and progression of several diseases. Microarray is the standard method for analyzing miRNA expression profiles; however, it has several disadvantages, including its limited detection of miRNAs. In recent years, advances in genome sequencing have led to the development of next-generation sequencing (NGS) technologies, which significantly advance genome sequencing speed and discovery. In this study, we compared the expression profiles obtained by next generation sequencing (NGS) with the profiles created using microarray to assess if NGS could produce a more accurate and complete miRNA profile. Total RNA from 14 hepatocellular carcinoma tumors (HCC) and 6 matched non-tumor control tissues were sequenced with Illumina MiSeq 50-bp single-end reads. Micro RNA expression profiles were estimated using miRDeep2 software. As a comparison, miRNA expression profiles for 11 out of 14 HCCs were also established by microarray (Agilent human microRNA microarray). The average total sequencing exceeded 2.2 million reads per sample and of those reads, approximately 57% mapped to the human genome. The average correlation for miRNA expression between microarray and NGS and subtraction were 0.613 and 0.587, respectively, while miRNA expression between technical replicates was 0.976. The diagnostic accuracy of HCC, p-value, and AUC were 90.0%, 7.22×10⁻⁴, and 0.92, respectively. In summary, NGS created an miRNA expression profile that was reproducible and comparable to that produced by microarray. Moreover, NGS discovered novel miRNAs that were otherwise undetectable by microarray. We believe that miRNA expression profiling by NGS can be a useful diagnostic tool applicable to multiple fields of medicine.     

Single-cell Iso-Sequencing enables rapid genome annotation for scRNAseq analysis

Single-cell RNA sequencing is a powerful technique that continues to expand across various biological applications. However, incomplete 3′-UTR annotations can impede single-cell analysis resulting in genes that are partially or completely uncounted. Performing single-cell RNA sequencing with incomplete 3′-UTR annotations can hinder the identification of cell identities and gene expression patterns and lead to erroneous biological inferences. We demonstrate that performing single-cell isoform sequencing in tandem with single-cell RNA sequencing can rapidly improve 3′-UTR annotations. Using threespine stickleback fish (Gasterosteus aculeatus), we show that gene models resulting from a minimal embryonic single-cell isoform sequencing dataset retained 26.1% greater single-cell RNA sequencing reads than gene models from Ensembl alone. Furthermore, pooling our single-cell sequencing isoforms with a previously published adult bulk Iso-Seq dataset from stickleback, and merging the annotation with the Ensembl gene models, resulted in a marginal improvement (+0.8%) over the single-cell isoform sequencing only dataset. In addition, isoforms identified by single-cell isoform sequencing included thousands of new splicing variants. The improved gene models obtained using single-cell isoform sequencing led to successful identification of cell types and increased the reads identified of many genes in our single-cell RNA sequencing stickleback dataset. Our work illuminates single-cell isoform sequencing as a cost-effective and efficient mechanism to rapidly annotate genomes for single-cell RNA sequencing.     

Ontology-Based Meta-Analysis of Global Collections of High-Throughput Public Data

Background

The investigation of the interconnections between the molecular and genetic events that govern biological systems is essential if we are to understand the development of disease and design effective novel treatments. Microarray and next-generation sequencing technologies have the potential to provide this information. However, taking full advantage of these approaches requires that biological connections be made across large quantities of highly heterogeneous genomic datasets. Leveraging the increasingly huge quantities of genomic data in the public domain is fast becoming one of the key challenges in the research community today.

Methodology/Results

We have developed a novel data mining framework that enables researchers to use this growing collection of public high-throughput data to investigate any set of genes or proteins. The connectivity between molecular states across thousands of heterogeneous datasets from microarrays and other genomic platforms is determined through a combination of rank-based enrichment statistics, meta-analyses, and biomedical ontologies. We address data quality concerns through dataset replication and meta-analysis and ensure that the majority of the findings are derived using multiple lines of evidence. As an example of our strategy and the utility of this framework, we apply our data mining approach to explore the biology of brown fat within the context of the thousands of publicly available gene expression datasets.

Conclusions

Our work presents a practical strategy for organizing, mining, and correlating global collections of large-scale genomic data to explore normal and disease biology. Using a hypothesis-free approach, we demonstrate how a data-driven analysis across very large collections of genomic data can reveal novel discoveries and evidence to support existing hypothesis.     

Challenges in studying genomic structural variant formation mechanisms: the short-read dilemma and beyond

Next-generation sequencing (NGS) technologies have revolutionised the analysis of genomic structural variants (SVs), providing significant insights into SV de novo formation based on analyses of rearrangement breakpoint junctions. The short DNA reads generated by NGS, however, have also created novel obstacles by biasing the ascertainment of SVs, an aspect that we refer to as the 'short-read dilemma'. For example, recent studies have found that SVs are often complex, with SV formation generating large numbers of breakpoints in a single event (multi-breakpoint SVs) or structurally polymorphic loci having multiple allelic states (multi-allelic SVs). This complexity may be obscured in short reads, unless the data is analysed and interpreted within its wider genomic context. We discuss how novel approaches will help to overcome the short-read dilemma, and how integration of other sources of information, including the structure of chromatin, may help in the future to deepen the understanding of SV formation processes.     

A comparative study of <Emphasis Type="Italic">k</Emphasis>-spectrum-based error correction methods for next-generation sequencing data analysis

Background

Innumerable opportunities for new genomic research have been stimulated by advancement in high-throughput next-generation sequencing (NGS). However, the pitfall of NGS data abundance is the complication of distinction between true biological variants and sequence error alterations during downstream analysis. Many error correction methods have been developed to correct erroneous NGS reads before further analysis, but independent evaluation of the impact of such dataset features as read length, genome size, and coverage depth on their performance is lacking. This comparative study aims to investigate the strength and weakness as well as limitations of some newest k-spectrum-based methods and to provide recommendations for users in selecting suitable methods with respect to specific NGS datasets.

Methods

Six k-spectrum-based methods, i.e., Reptile, Musket, Bless, Bloocoo, Lighter, and Trowel, were compared using six simulated sets of paired-end Illumina sequencing data. These NGS datasets varied in coverage depth (10× to 120×), read length (36 to 100 bp), and genome size (4.6 to 143 MB). Error Correction Evaluation Toolkit (ECET) was employed to derive a suite of metrics (i.e., true positives, false positive, false negative, recall, precision, gain, and F-score) for assessing the correction quality of each method.

Results

Results from computational experiments indicate that Musket had the best overall performance across the spectra of examined variants reflected in the six datasets. The lowest accuracy of Musket (F-score?=?0.81) occurred to a dataset with a medium read length (56 bp), a medium coverage (50×), and a small-sized genome (5.4 MB). The other five methods underperformed (F-score?<?0.80) and/or failed to process one or more datasets.

Conclusions

This study demonstrates that various factors such as coverage depth, read length, and genome size may influence performance of individual k-spectrum-based error correction methods. Thus, efforts have to be paid in choosing appropriate methods for error correction of specific NGS datasets. Based on our comparative study, we recommend Musket as the top choice because of its consistently superior performance across all six testing datasets. Further extensive studies are warranted to assess these methods using experimental datasets generated by NGS platforms (e.g., 454, SOLiD, and Ion Torrent) under more diversified parameter settings (k-mer values and edit distances) and to compare them against other non-k-spectrum-based classes of error correction methods.

     

Prevention,diagnosis and treatment of high‐throughput sequencing data pathologies

High‐throughput sequencing (HTS) technologies generate millions of sequence reads from DNA/RNA molecules rapidly and cost‐effectively, enabling single investigator laboratories to address a variety of ‘omics’ questions in nonmodel organisms, fundamentally changing the way genomic approaches are used to advance biological research. One major challenge posed by HTS is the complexity and difficulty of data quality control (QC). While QC issues associated with sample isolation, library preparation and sequencing are well known and protocols for their handling are widely available, the QC of the actual sequence reads generated by HTS is often overlooked. HTS‐generated sequence reads can contain various errors, biases and artefacts whose identification and amelioration can greatly impact subsequent data analysis. However, a systematic survey on QC procedures for HTS data is still lacking. In this review, we begin by presenting standard ‘health check‐up’ QC procedures recommended for HTS data sets and establishing what ‘healthy’ HTS data look like. We next proceed by classifying errors, biases and artefacts present in HTS data into three major types of ‘pathologies’, discussing their causes and symptoms and illustrating with examples their diagnosis and impact on downstream analyses. We conclude this review by offering examples of successful ‘treatment’ protocols and recommendations on standard practices and treatment options. Notwithstanding the speed with which HTS technologies – and consequently their pathologies – change, we argue that careful QC of HTS data is an important – yet often neglected – aspect of their application in molecular ecology, and lay the groundwork for developing a HTS data QC ‘best practices’ guide.     

First insights into the metagenome of Egyptian mummies using next-generation sequencing

We applied, for the first time, next-generation sequencing (NGS) technology on Egyptian mummies. Seven NGS datasets obtained from five randomly selected Third Intermediate to Graeco-Roman Egyptian mummies (806 BC–124AD) and two unearthed pre-contact Bolivian lowland skeletons were generated and characterised. The datasets were contrasted to three recently published NGS datasets obtained from cold-climate regions, i.e. the Saqqaq, the Denisova hominid and the Alpine Iceman. Analysis was done using one million reads of each newly generated or published dataset. Blastn and megablast results were analysed using MEGAN software. Distinct NGS results were replicated by specific and sensitive polymerase chain reaction (PCR) protocols in ancient DNA dedicated laboratories. Here, we provide unambiguous identification of authentic DNA in Egyptian mummies. The NGS datasets showed variable contents of endogenous DNA harboured in tissues. Three of five mummies displayed a human DNA proportion comparable to the human read count of the Saqqaq permafrost-preserved specimen. Furthermore, a metagenomic signature unique to mummies was displayed. By applying a “bacterial fingerprint”, discrimination among mummies and other remains from warm areas outside Egypt was possible. Due to the absence of an adequate environment monitoring, a bacterial bloom was identified when analysing different biopsies from the same mummies taken after a lapse of time of 1.5 years. Plant kingdom representation in all mummy datasets was unique and could be partially associated with their use in embalming materials. Finally, NGS data showed the presence of Plasmodium falciparum and Toxoplasma gondii DNA sequences, indicating malaria and toxoplasmosis in these mummies. We demonstrate that endogenous ancient DNA can be extracted from mummies and serve as a proper template for the NGS technique, thus, opening new pathways of investigation for future genome sequencing of ancient Egyptian individuals.     

Current challenges and solutions of de novo assembly

Background: Next-generation sequencing (NGS) technologies have fostered an unprecedented proliferation of high-throughput sequencing projects and a concomitant development of novel algorithms for the assembly of short reads. However, numerous technical or computational challenges in de novo assembly still remain, although many new ideas and solutions have been suggested to tackle the challenges in both experimental and computational settings.Results: In this review, we first briefly introduce some of the major challenges faced by NGS sequence assembly. Then, we analyze the characteristics of various sequencing platforms and their impact on assembly results. After that, we classify de novo assemblers according to their frameworks (overlap graph-based, de Bruijn graph-based and string graph-based), and introduce the characteristics of each assembly tool and their adaptation scene. Next, we introduce in detail the solutions to the main challenges of de novo assembly of next generation sequencing data, single-cell sequencing data and single molecule sequencing data. At last, we discuss the application of SMS long reads in solving problems encountered in NGS assembly.Conclusions: This review not only gives an overview of the latest methods and developments in assembly algorithms, but also provides guidelines to determine the optimal assembly algorithm for a given input sequencing data type.     

Next generation sequencing technology: Advances and applications

Impressive progress has been made in the field of Next Generation Sequencing (NGS). Through advancements in the fields of molecular biology and technical engineering, parallelization of the sequencing reaction has profoundly increased the total number of produced sequence reads per run. Current sequencing platforms allow for a previously unprecedented view into complex mixtures of RNA and DNA samples. NGS is currently evolving into a molecular microscope finding its way into virtually every fields of biomedical research. In this chapter we review the technical background of the different commercially available NGS platforms with respect to template generation and the sequencing reaction and take a small step towards what the upcoming NGS technologies will bring. We close with an overview of different implementations of NGS into biomedical research. This article is part of a Special Issue entitled: From Genome to Function.     

Library preparation methods for next-generation sequencing: Tone down the bias

Next-generation sequencing (NGS) has caused a revolution in biology. NGS requires the preparation of libraries in which (fragments of) DNA or RNA molecules are fused with adapters followed by PCR amplification and sequencing. It is evident that robust library preparation methods that produce a representative, non-biased source of nucleic acid material from the genome under investigation are of crucial importance. Nevertheless, it has become clear that NGS libraries for all types of applications contain biases that compromise the quality of NGS datasets and can lead to their erroneous interpretation. A detailed knowledge of the nature of these biases will be essential for a careful interpretation of NGS data on the one hand and will help to find ways to improve library quality or to develop bioinformatics tools to compensate for the bias on the other hand. In this review we discuss the literature on bias in the most common NGS library preparation protocols, both for DNA sequencing (DNA-seq) as well as for RNA sequencing (RNA-seq). Strikingly, almost all steps of the various protocols have been reported to introduce bias, especially in the case of RNA-seq, which is technically more challenging than DNA-seq. For each type of bias we discuss methods for improvement with a view to providing some useful advice to the researcher who wishes to convert any kind of raw nucleic acid into an NGS library.     

BM-map: Bayesian mapping of multireads for next-generation sequencing data

Next-generation sequencing (NGS) technology generates millions of short reads, which provide valuable information for various aspects of cellular activities and biological functions. A key step in NGS applications (e.g., RNA-Seq) is to map short reads to correct genomic locations within the source genome. While most reads are mapped to a unique location, a significant proportion of reads align to multiple genomic locations with equal or similar numbers of mismatches; these are called multireads. The ambiguity in mapping the multireads may lead to bias in downstream analyses. Currently, most practitioners discard the multireads in their analysis, resulting in a loss of valuable information, especially for the genes with similar sequences. To refine the read mapping, we develop a Bayesian model that computes the posterior probability of mapping a multiread to each competing location. The probabilities are used for downstream analyses, such as the quantification of gene expression. We show through simulation studies and RNA-Seq analysis of real life data that the Bayesian method yields better mapping than the current leading methods. We provide a C++ program for downloading that is being packaged into a user-friendly software.     

Hybrid assembly of ultra-long Nanopore reads augmented with 10x-Genomics contigs: Demonstrated with a human genome

The 3rd generation of sequencing (3GS) technologies generate ultra-long reads (up to 1 Mb), which makes it possible to eliminate gaps and effectively resolve repeats in genome assembly. However, the 3GS technologies suffer from the high base-level error rates (15%–40%) and high sequencing costs. To address these issues, the hybrid assembly strategy, which utilizes both 3GS reads and inexpensive NGS (next generation sequencing) short reads, was invented. Here, we use 10×-Genomics® technology, which integrates a novel bar-coding strategy with Illumina® NGS with an advantage of revealing long-range sequence information, to replace common NGS short reads for hybrid assembly of long erroneous 3GS reads. We demonstrate the feasibility of integrating the 3GS with 10×-Genomics technologies for a new strategy of hybrid de novo genome assembly by utilizing DBG2OLC and Sparc software packages, previously developed by the authors for regular hybrid assembly. Using a human genome as an example, we show that with only 7× coverage of ultra-long Nanopore® reads, augmented with 10× reads, our approach achieved nearly the same level of quality, compared with non-hybrid assembly with 35× coverage of Nanopore reads. Compared with the assembly with 10×-Genomics reads alone, our assembly is gapless with slightly high cost. These results suggest that our new hybrid assembly with ultra-long 3GS reads augmented with 10×-Genomics reads offers a low-cost (less than ¼ the cost of the non-hybrid assembly) and computationally light-weighted (only took 109 calendar hours with peak memory-usage = 61GB on a dual-CPU office workstation) solution for extending the wide applications of the 3GS technologies.     

Connectivity Mapping for Candidate Therapeutics Identification Using Next Generation Sequencing RNA-Seq Data

The advent of next generation sequencing technologies (NGS) has expanded the area of genomic research, offering high coverage and increased sensitivity over older microarray platforms. Although the current cost of next generation sequencing is still exceeding that of microarray approaches, the rapid advances in NGS will likely make it the platform of choice for future research in differential gene expression. Connectivity mapping is a procedure for examining the connections among diseases, genes and drugs by differential gene expression initially based on microarray technology, with which a large collection of compound-induced reference gene expression profiles have been accumulated. In this work, we aim to test the feasibility of incorporating NGS RNA-Seq data into the current connectivity mapping framework by utilizing the microarray based reference profiles and the construction of a differentially expressed gene signature from a NGS dataset. This would allow for the establishment of connections between the NGS gene signature and those microarray reference profiles, alleviating the associated incurring cost of re-creating drug profiles with NGS technology. We examined the connectivity mapping approach on a publicly available NGS dataset with androgen stimulation of LNCaP cells in order to extract candidate compounds that could inhibit the proliferative phenotype of LNCaP cells and to elucidate their potential in a laboratory setting. In addition, we also analyzed an independent microarray dataset of similar experimental settings. We found a high level of concordance between the top compounds identified using the gene signatures from the two datasets. The nicotine derivative cotinine was returned as the top candidate among the overlapping compounds with potential to suppress this proliferative phenotype. Subsequent lab experiments validated this connectivity mapping hit, showing that cotinine inhibits cell proliferation in an androgen dependent manner. Thus the results in this study suggest a promising prospect of integrating NGS data with connectivity mapping.     

Identification of Optimum Sequencing Depth Especially for De Novo Genome Assembly of Small Genomes Using Next Generation Sequencing Data

Next Generation Sequencing (NGS) is a disruptive technology that has found widespread acceptance in the life sciences research community. The high throughput and low cost of sequencing has encouraged researchers to undertake ambitious genomic projects, especially in de novo genome sequencing. Currently, NGS systems generate sequence data as short reads and de novo genome assembly using these short reads is computationally very intensive. Due to lower cost of sequencing and higher throughput, NGS systems now provide the ability to sequence genomes at high depth. However, currently no report is available highlighting the impact of high sequence depth on genome assembly using real data sets and multiple assembly algorithms. Recently, some studies have evaluated the impact of sequence coverage, error rate and average read length on genome assembly using multiple assembly algorithms, however, these evaluations were performed using simulated datasets. One limitation of using simulated datasets is that variables such as error rates, read length and coverage which are known to impact genome assembly are carefully controlled. Hence, this study was undertaken to identify the minimum depth of sequencing required for de novo assembly for different sized genomes using graph based assembly algorithms and real datasets. Illumina reads for E.coli (4.6 MB) S.kudriavzevii (11.18 MB) and C.elegans (100 MB) were assembled using SOAPdenovo, Velvet, ABySS, Meraculous and IDBA-UD. Our analysis shows that 50X is the optimum read depth for assembling these genomes using all assemblers except Meraculous which requires 100X read depth. Moreover, our analysis shows that de novo assembly from 50X read data requires only 6–40 GB RAM depending on the genome size and assembly algorithm used. We believe that this information can be extremely valuable for researchers in designing experiments and multiplexing which will enable optimum utilization of sequencing as well as analysis resources.