MetaVelvet: an extension of Velvet assembler to de novo metagenome assembly from short sequence reads

An important step in ‘metagenomics’ analysis is the assembly of multiple genomes from mixed sequence reads of multiple species in a microbial community. Most conventional pipelines use a single-genome assembler with carefully optimized parameters. A limitation of a single-genome assembler for de novo metagenome assembly is that sequences of highly abundant species are likely misidentified as repeats in a single genome, resulting in a number of small fragmented scaffolds. We extended a single-genome assembler for short reads, known as ‘Velvet’, to metagenome assembly, which we called ‘MetaVelvet’, for mixed short reads of multiple species. Our fundamental concept was to first decompose a de Bruijn graph constructed from mixed short reads into individual sub-graphs, and second, to build scaffolds based on each decomposed de Bruijn sub-graph as an isolate species genome. We made use of two features, the coverage (abundance) difference and graph connectivity, for the decomposition of the de Bruijn graph. For simulated datasets, MetaVelvet succeeded in generating significantly higher N50 scores than any single-genome assemblers. MetaVelvet also reconstructed relatively low-coverage genome sequences as scaffolds. On real datasets of human gut microbial read data, MetaVelvet produced longer scaffolds and increased the number of predicted genes.     

Targeted assembly of short sequence reads

As next-generation sequence (NGS) production continues to increase, analysis is becoming a significant bottleneck. However, in situations where information is required only for specific sequence variants, it is not necessary to assemble or align whole genome data sets in their entirety. Rather, NGS data sets can be mined for the presence of sequence variants of interest by localized assembly, which is a faster, easier, and more accurate approach. We present TASR, a streamlined assembler that interrogates very large NGS data sets for the presence of specific variants by only considering reads within the sequence space of input target sequences provided by the user. The NGS data set is searched for reads with an exact match to all possible short words within the target sequence, and these reads are then assembled stringently to generate a consensus of the target and flanking sequence. Typically, variants of a particular locus are provided as different target sequences, and the presence of the variant in the data set being interrogated is revealed by a successful assembly outcome. However, TASR can also be used to find unknown sequences that flank a given target. We demonstrate that TASR has utility in finding or confirming genomic mutations, polymorphisms, fusions and integration events. Targeted assembly is a powerful method for interrogating large data sets for the presence of sequence variants of interest. TASR is a fast, flexible and easy to use tool for targeted assembly.     

Whole-genome re-sequencing of non-model organisms: lessons from unmapped reads

Unmapped reads are often discarded from the analysis of whole-genome re-sequencing, but new biological information and insights can be uncovered through their analysis. In this paper, we investigate unmapped reads from the re-sequencing data of 33 pea aphid genomes from individuals specialized on different host plants. The unmapped reads for each individual were retrieved following mapping to the Acyrthosiphon pisum reference genome and its mitochondrial and symbiont genomes. These sets of unmapped reads were then cross-compared, revealing that a significant number of these unmapped sequences were conserved across individuals. Interestingly, sequences were most commonly shared between individuals adapted to the same host plant, suggesting that these sequences may contribute to the divergence between host plant specialized biotypes. Analysis of the contigs obtained from assembling the unmapped reads pooled by biotype allowed us to recover some divergent genomic regions previously excluded from analysis and to discover putative novel sequences of A. pisum and its symbionts. In conclusion, this study emphasizes the interest of the unmapped component of re-sequencing data sets and the potential loss of important information. We here propose strategies to aid the capture and interpretation of this information.     

Structural variation in two human genomes mapped at single-nucleotide resolution by whole genome de novo assembly

Here we use whole-genome de novo assembly of second-generation sequencing reads to map structural variation (SV) in an Asian genome and an African genome. Our approach identifies small- and intermediate-size homozygous variants (1-50 kb) including insertions, deletions, inversions and their precise breakpoints, and in contrast to other methods, can resolve complex rearrangements. In total, we identified 277,243 SVs ranging in length from 1-23 kb. Validation using computational and experimental methods suggests that we achieve overall <6% false-positive rate and <10% false-negative rate in genomic regions that can be assembled, which outperforms other methods. Analysis of the SVs in the genomes of 106 individuals sequenced as part of the 1000 Genomes Project suggests that SVs account for a greater fraction of the diversity between individuals than do single-nucleotide polymorphisms (SNPs). These findings demonstrate that whole-genome de novo assembly is a feasible approach to deriving more comprehensive maps of genetic variation.     

Multiple sequence assembly from reads alignable to a common reference genome

We describe a set of computational problems motivated by certain analysis tasks in genome resequencing. These are assembly problems for which multiple distinct sequences must be assembled, but where the relative positions of reads to be assembled are already known. This information is obtained from a common reference genome and is characteristic of resequencing experiments. The simplest variant of the problem aims at determining a minimum set of superstrings such that each sequenced read matches at least one superstring. We give an algorithm with time complexity O(N), where N is the sum of the lengths of reads, substantially improving on previous algorithms for solving the same problem. We also examine the problem of finding the smallest number of reads to remove such that the remaining reads are consistent with k superstrings. By exploiting a surprising relationship with the minimum cost flow problem, we show that this problem can be solved in polynomial time when nested reads are excluded. If nested reads are permitted, this problem of removing the minimum number of reads becomes NP-hard. We show that permitting mismatches between reads and their nearest superstrings generally renders these problems NP-hard.     

Meraculous: de novo genome assembly with short paired-end reads

We describe a new algorithm, meraculous, for whole genome assembly of deep paired-end short reads, and apply it to the assembly of a dataset of paired 75-bp Illumina reads derived from the 15.4 megabase genome of the haploid yeast Pichia stipitis. More than 95% of the genome is recovered, with no errors; half the assembled sequence is in contigs longer than 101 kilobases and in scaffolds longer than 269 kilobases. Incorporating fosmid ends recovers entire chromosomes. Meraculous relies on an efficient and conservative traversal of the subgraph of the k-mer (deBruijn) graph of oligonucleotides with unique high quality extensions in the dataset, avoiding an explicit error correction step as used in other short-read assemblers. A novel memory-efficient hashing scheme is introduced. The resulting contigs are ordered and oriented using paired reads separated by ～280 bp or ～3.2 kbp, and many gaps between contigs can be closed using paired-end placements. Practical issues with the dataset are described, and prospects for assembling larger genomes are discussed.     

Feature-by-feature--evaluating de novo sequence assembly

The whole-genome sequence assembly (WGSA) problem is among one of the most studied problems in computational biology. Despite the availability of a plethora of tools (i.e., assemblers), all claiming to have solved the WGSA problem, little has been done to systematically compare their accuracy and power. Traditional methods rely on standard metrics and read simulation: while on the one hand, metrics like N50 and number of contigs focus only on size without proportionately emphasizing the information about the correctness of the assembly, comparisons performed on simulated dataset, on the other hand, can be highly biased by the non-realistic assumptions in the underlying read generator. Recently the Feature Response Curve (FRC) method was proposed to assess the overall assembly quality and correctness: FRC transparently captures the trade-offs between contigs' quality against their sizes. Nevertheless, the relationship among the different features and their relative importance remains unknown. In particular, FRC cannot account for the correlation among the different features. We analyzed the correlation among different features in order to better describe their relationships and their importance in gauging assembly quality and correctness. In particular, using multivariate techniques like principal and independent component analysis we were able to estimate the "excess-dimensionality" of the feature space. Moreover, principal component analysis allowed us to show how poorly the acclaimed N50 metric describes the assembly quality. Applying independent component analysis we identified a subset of features that better describe the assemblers performances. We demonstrated that by focusing on a reduced set of highly informative features we can use the FRC curve to better describe and compare the performances of different assemblers. Moreover, as a by-product of our analysis, we discovered how often evaluation based on simulated data, obtained with state of the art simulators, lead to not-so-realistic results.     

Local de novo assembly of RAD paired-end contigs using short sequencing reads

Despite the power of massively parallel sequencing platforms, a drawback is the short length of the sequence reads produced. We demonstrate that short reads can be locally assembled into longer contigs using paired-end sequencing of restriction-site associated DNA (RAD-PE) fragments. We use this RAD-PE contig approach to identify single nucleotide polymorphisms (SNPs) and determine haplotype structure in threespine stickleback and to sequence E. coli and stickleback genomic DNA with overlapping contigs of several hundred nucleotides. We also demonstrate that adding a circularization step allows the local assembly of contigs up to 5 kilobases (kb) in length. The ease of assembly and accuracy of the individual contigs produced from each RAD site sequence suggests RAD-PE sequencing is a useful way to convert genome-wide short reads into individually-assembled sequences hundreds or thousands of nucleotides long.     

Efficient de novo assembly of single-cell bacterial genomes from short-read data sets

Whole genome amplification by the multiple displacement amplification (MDA) method allows sequencing of DNA from single cells of bacteria that cannot be cultured. Assembling a genome is challenging, however, because MDA generates highly nonuniform coverage of the genome. Here we describe an algorithm tailored for short-read data from single cells that improves assembly through the use of a progressively increasing coverage cutoff. Assembly of reads from single Escherichia coli and Staphylococcus aureus cells captures >91% of genes within contigs, approaching the 95% captured from an assembly based on many E. coli cells. We apply this method to assemble a genome from a single cell of an uncultivated SAR324 clade of Deltaproteobacteria, a cosmopolitan bacterial lineage in the global ocean. Metabolic reconstruction suggests that SAR324 is aerobic, motile and chemotaxic. Our approach enables acquisition of genome assemblies for individual uncultivated bacteria using only short reads, providing cell-specific genetic information absent from metagenomic studies.     

Compression of next-generation sequencing reads aided by highly efficient de novo assembly

We present Quip, a lossless compression algorithm for next-generation sequencing data in the FASTQ and SAM/BAM formats. In addition to implementing reference-based compression, we have developed, to our knowledge, the first assembly-based compressor, using a novel de novo assembly algorithm. A probabilistic data structure is used to dramatically reduce the memory required by traditional de Bruijn graph assemblers, allowing millions of reads to be assembled very efficiently. Read sequences are then stored as positions within the assembled contigs. This is combined with statistical compression of read identifiers, quality scores, alignment information and sequences, effectively collapsing very large data sets to <15% of their original size with no loss of information. Availability: Quip is freely available under the 3-clause BSD license from http://cs.washington.edu/homes/dcjones/quip.     

The genome of flax (Linum usitatissimum) assembled de novo from short shotgun sequence reads

Flax (Linum usitatissimum) is an ancient crop that is widely cultivated as a source of fiber, oil and medicinally relevant compounds. To accelerate crop improvement, we performed whole‐genome shotgun sequencing of the nuclear genome of flax. Seven paired‐end libraries ranging in size from 300 bp to 10 kb were sequenced using an Illumina genome analyzer. A de novo assembly, comprised exclusively of deep‐coverage (approximately 94× raw, approximately 69× filtered) short‐sequence reads (44–100 bp), produced a set of scaffolds with N₅₀ = 694 kb, including contigs with N₅₀ = 20.1 kb. The contig assembly contained 302 Mb of non‐redundant sequence representing an estimated 81% genome coverage. Up to 96% of published flax ESTs aligned to the whole‐genome shotgun scaffolds. However, comparisons with independently sequenced BACs and fosmids showed some mis‐assembly of regions at the genome scale. A total of 43 384 protein‐coding genes were predicted in the whole‐genome shotgun assembly, and up to 93% of published flax ESTs, and 86% of A. thaliana genes aligned to these predicted genes, indicating excellent coverage and accuracy at the gene level. Analysis of the synonymous substitution rates (K_s) observed within duplicate gene pairs was consistent with a recent (5–9 MYA) whole‐genome duplication in flax. Within the predicted proteome, we observed enrichment of many conserved domains (Pfam‐A) that may contribute to the unique properties of this crop, including agglutinin proteins. Together these results show that de novo assembly, based solely on whole‐genome shotgun short‐sequence reads, is an efficient means of obtaining nearly complete genome sequence information for some plant species.     

De novo assembly of the Pseudomonas syringae pv. syringae B728a genome using Illumina/Solexa short sequence reads

Illumina's Genome Analyzer generates ultra-short sequence reads, typically 36 nucleotides in length, and is primarily intended for resequencing. We tested the potential of this technology for de novo sequence assembly on the 6 Mbp genome of Pseudomonas syringae pv. syringae B728a with several freely available assembly software packages. Using an unpaired data set, velvet assembled >96% of the genome into contigs with an N50 length of 8289 nucleotides and an error rate of 0.33%. edena generated smaller contigs (N50 was 4192 nucleotides) and comparable error rates. ssake and vcake yielded shorter contigs with very high error rates. Assembly of paired-end sequence data carrying 400 bp inserts produced longer contigs (N50 up to 15 628 nucleotides), but with increased error rates (0.5%). Contig length and error rate were very sensitive to the choice of parameter values. Noncoding RNA genes were poorly resolved in de novo assemblies, while >90% of the protein-coding genes were assembled with 100% accuracy over their full length. This study demonstrates that, in practice, de novo assembly of 36-nucleotide reads can generate reasonably accurate assemblies from about 40 × deep sequence data sets. These draft assemblies are useful for exploring an organism's proteomic potential, at a very economic low cost.     

A biologist's guide to de novo genome assembly using next-generation sequence data: A test with fungal genomes

We offer a guide to de novo genome assembly¹ using sequence data generated by the Illumina platform for biologists working with fungi or other organisms whose genomes are less than 100 Mb in size. The guide requires no familiarity with sequencing assembly technology or associated computer programs. It defines commonly used terms in genome sequencing and assembly; provides examples of assembling short-read genome sequence data for four strains of the fungus Grosmannia clavigera using four assembly programs; gives examples of protocols and software; and presents a commented flowchart that extends from DNA preparation for submission to a sequencing center, through to processing and assembly of the raw sequence reads using freely available operating systems and software.     

SNPing variation from genomes

A report on the fourth International Meeting on Single Nucleotide Polymorphisms and Complex Genome Analysis, Stockholm, Sweden, 10-13 October 2001.     

StrainXpress: strain aware metagenome assembly from short reads

Next-generation sequencing–based metagenomics has enabled to identify microorganisms in characteristic habitats without the need for lengthy cultivation. Importantly, clinically relevant phenomena such as resistance to medication, virulence or interactions with the environment can vary already within species. Therefore, a major current challenge is to reconstruct individual genomes from the sequencing reads at the level of strains, and not just the level of species. However, strains of one species can differ only by minor amounts of variants, which makes it difficult to distinguish them. Despite considerable recent progress, related approaches have remained fragmentary so far. Here, we present StrainXpress, as a comprehensive solution to the problem of strain aware metagenome assembly from next-generation sequencing reads. In experiments, StrainXpress reconstructs strain-specific genomes from metagenomes that involve up to >1000 strains and proves to successfully deal with poorly covered strains. The amount of reconstructed strain-specific sequence exceeds that of the current state-of-the-art approaches by on average 26.75% across all data sets (first quartile: 18.51%, median: 26.60%, third quartile: 35.05%).     

Simultaneous alignment of short reads against multiple genomes

Sequencing-by-synthesis technologies can reduce the cost of generating de novo genome assemblies. We report a method for assembling draft genome sequences of eukaryotic organisms that integrates sequence information from different sources, and demonstrate its effectiveness by assembling an approximately 32.5 Mb draft genome sequence for the forest pathogen Grosmannia clavigera, an ascomycete fungus. We also developed a method for assessing draft assemblies using Illumina paired end read data and demonstrate how we are using it to guide future sequence finishing. Our results demonstrate that eukaryotic genome sequences can be accurately assembled by combining Illumina, 454 and Sanger sequence data.