Evaluating assembly and variant calling software for strain-resolved analysis of large DNA-viruses

Deng, Zhi-Luo; Dhingra, Akshay; Fritz, Adrian; Götting, Jasper; Muench, Philipp; Steinbrueck, Lars; Schulz, Thomas F.; Ganzenmueller, Tina; McHardy, Alice C.

doi:10.1101/2020.05.14.095265

Cited by 7 publications

(21 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Assemblies were evaluated using metaQUAST v.5.0.2 with the benchmarking workflow QuasiModo 14 , based on common assembly metrics, the composite assembly score, recall and precision in strain-resolved genome assembly, as before, and the top performing methods falling in the 95-100% range of results identified for every metric.…”

Section: Hiv-3 In Silico Mixturementioning

confidence: 99%

“…For the Wisconsin strains, which were passaged for up to two rounds in cell cultures, the reconstructed minor strains from short read data had more evolutionary divergences. In comparison to calls from the variant caller Lofreq 41 , (Supplementary Methods), which performs particularly well on mixed strain viral data 14 , both identified 17 (65.4%) of overall 26 identified variant sites (mutations and up to 2bp indels). Interestingly, most of these are C->T transitions, indicating a tendency to alter genome composition 42 (Supplementary Table S6).…”

Section: After Correcting For Pcr Amplification and Sequencing Errorsmentioning

confidence: 99%

“…The evaluation was performed using metaQUAST 50 v.5.0.2, which is commonly used to evaluate metagenome assemblies and provides useful metrics for measuring completeness (genome fraction), continuity (NGA50, largest alignment) and accuracy (mismatches per 100kb, duplication ratio) of assemblies and has specific options for analyzing strain-resolved assemblies. In addition, we calculated metrics for assessing strain-resolved assembly; the strain recall, specifying the fraction of correctly assembled strains (more than 90 (80)% genome fraction and less than 1 (5) mismatches/kb), the strain precision, specifying the fraction of correctly assembled strain genomes of all provided genome assemblies (true positives defined as in recall; total number of genome assemblies estimated as number of ground truth genomes with at least one mapping contig * duplication ratio), as well as the composite assembly quality score, we previously defined 14 other metrics and then weighs with a weight of 0.3 for genome fraction and largest alignment, respectively and a weight of 0.1 for the other metrics.…”

Section: Performance Evaluationmentioning

confidence: 99%

“…Of all evaluated de novo assemblers, Haploflow performed best across all metrics and the composite assembly score ( Figure S2 ) , assembling all three strains almost completely (more than 90%), with less than 1 mismatch/kb, providing no false positive strain assemblies -that for some methods (QuasiRecomb) reached several thousand strains -and with more than double the assembly contiguity (NGA50) than the second best method (PEHaplo) 14 or were terminated after ten days (Haploflow on average required 1.5h per sample).…”

Section: Hiv-3 In Silico Mixturementioning

confidence: 99%

“…Currently there are predominantly two types of methods for this problem, viral haplotype assemblers 6,7 and general (meta)genome assemblers [8][9][10][11][12] . Assembly of individual strains is very difficult, especially if variation is low and few reads span varying sites, resulting in highly fragmented strain genome reconstructions or consensus assemblies 13,14 .…”

mentioning

confidence: 99%

See 4 more Smart Citations

Haploflow: Strain-resolvedde novoassembly of viral genomes

Fritz

Bremges

Deng

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In viral infections often multiple related viral strains are present, due to coinfection or within-host evolution. We describe Haploflow, a de Bruijn graph-based assembler for de novo genome assembly of viral strains from mixed sequence samples using a novel flow algorithm. We assessed Haploflow across multiple benchmark data sets of increasing complexity, showing that Haploflow is faster and more accurate than viral haplotype assemblers and generic metagenome assemblers not aiming to reconstruct strains. Haplotype reconstructed high-quality strain-resolved assemblies from clinical HCMV samples and SARS-CoV-2 genomes from wastewater metagenomes identical to genomes from clinical isolates.

show abstract

Section: Hiv-3 In Silico Mixturementioning

confidence: 99%

Section: After Correcting For Pcr Amplification and Sequencing Errorsmentioning

confidence: 99%

Section: Performance Evaluationmentioning

confidence: 99%

Section: Hiv-3 In Silico Mixturementioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Haploflow: Strain-resolvedde novoassembly of viral genomes

Fritz

Bremges

Deng

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Identifying high-confidence variants in human cytomegalovirus genomes sequenced from clinical samples

et al. 2022

View full text Add to dashboard Cite

Understanding the intrahost evolution of viral populations has implications in pathogenesis, diagnosis and treatment, and has recently made impressive advances from developments in high-throughput sequencing. However, the underlying analyses are very sensitive to sources of bias, error and artefact in the data, and it is important that these are addressed adequately if robust conclusions are to be drawn. The key factors include: (i) determining the number of viral strains present in the sample analysed; (ii) monitoring the extent to which the data represent these strains and assessing the quality of these data; (iii) dealing with the effects of cross-contamination; and (iv) ensuring that the results are reproducible. We investigated these factors by generating sequence datasets, including biological and technical replicates, directly from clinical samples obtained from a small cohort of patients who had been infected congenitally with the herpesvirus human cytomegalovirus, with the aim of developing a strategy for identifying high-confidence intrahost variants. We found that such variants were few in number and typically present in low proportions, and concluded that human cytomegalovirus exhibits a very low level of intrahost variability. In addition to clarifying the situation regarding human cytomegalovirus, our strategy has wider applicability to understanding the intrahost variability of other viruses.

show abstract

High-throughput engineering of cytoplasmic- and nuclear-replicating large dsDNA viruses by CRISPR/Cas9

López-Muñoz

Rastrojo

Martı́n

et al. 2022

Journal of General Virology

View full text Add to dashboard Cite

The application of CRISPR/Cas9 to improve genome engineering efficiency for large dsDNA viruses has been extensively described, but a robust and versatile method for high-throughput generation of marker-free recombinants for a desired locus has not yet been reported. Cytoplasmic-replicating viruses use their own repair enzymes for homologous recombination, while nuclear-replicating viruses use the host repair machinery. This is translated into a wide range of Cas9-induced homologous recombination efficiencies, depending on the virus replication compartment and viral/host repair machinery characteristics and accessibility. However, the use of Cas9 as a selection agent to target parental virus genomes robustly improves the selection of desired recombinants across large dsDNA viruses. We used ectromelia virus (ECTV) and herpes simplex virus (HSV) type 1 and 2 to optimize a CRISPR/Cas9 method that can be used versatilely for efficient genome editing and selection of both cytoplasmic- and nuclear-replicating viruses. We performed a genome-wide genetic variant analysis of mutations located at predicted off-target sequences for 20 different recombinants, showing off-target-free accuracy by deep sequencing. Our results support this optimized method as an efficient, accurate and versatile approach to enhance the two critical factors of high-throughput viral genome engineering: generation and colour-based selection of recombinants. This application of CRISPR/Cas9 reduces the time and labour for screening of desired recombinants, allowing for high-throughput generation of large collections of mutant dsDNA viruses for a desired locus, optimally in less than 2 weeks.

show abstract

Evaluating assembly and variant calling software for strain-resolved analysis of large DNA-viruses

Abstract: 1

Cited by 7 publications

References 59 publications

Haploflow: Strain-resolvedde novoassembly of viral genomes

Haploflow: Strain-resolvedde novoassembly of viral genomes

Identifying high-confidence variants in human cytomegalovirus genomes sequenced from clinical samples

High-throughput engineering of cytoplasmic- and nuclear-replicating large dsDNA viruses by CRISPR/Cas9

Contact Info

Product

Resources

About