Does Choice Matter? Reference-Based Alignment for Molecular Epidemiology of Tuberculosis

An imputed ancestral reference genome for the Mycobacterium tuberculosis complex better captures structural genomic diversity for reference-based alignment workflows

Reference-based alignment of short-reads is a widely used technique in genomic analysis of the Mycobacterium tuberculosis complex (MTBC) and the choice of reference sequence impacts the interpretation of analyses. The most widely used reference genomes include the ATCC type strain (H37Rv) and the putative MTBC ancestral sequence of Comas et al. both of which are based on a lineage 4 sequence. As such, these reference sequences do not capture all of the structural variation known to be present in the ancestor of the MTBC. To better represent the base of the MTBC, we generated an imputed ancestral genomic sequence, termed MTBC0 from reference-free alignments of closed MTBC genomes. When used as a reference sequence in alignment workflows, MTBC0 mapped more short sequencing reads and called more pairwise SNPs relative to the Comas et al. sequence while exhibiting minimal impact on the overall phylogeny of MTBC. The results also show that MTBC0 provides greater fidelity in capturing genomic variation and allows for the inclusion of regions absent from H37Rv in standard MTBC workflows without additional steps. The use of MTBC0 as an ancestral reference sequence in standard workflows modestly improved read mapping, SNP calling and intuitively facilitates the study of structural variation and evolution in MTBC.

Comparative genomic analyses of multi-drug resistant Mycobacterium tuberculosis isolates from Nepal and other geographical locations

Nepal exhibits a tuberculosis (TB) incidence rate that is comparable to neighbouring high TB incidence countries. In addition, it records >500 cases of multi-drug resistant (MDR) TB each year. The objective of this study was to perform whole-genome bioinformatic analysis on MDR-TB isolates from Nepal (n = 19) to identify the specific mutations underlying their phenotypic resistance. In addition, we examined the dominant genotype among the Nepal MDR-TB isolates, the East-Asian Beijing sub-lineage, to determine its relatedness to a panel of 1274 genomes of international strains available from public databases. These analyses provided evidence that the XDR-TB isolates in our collection were not derived from importation of primary XDR-TB to Nepal but were more likely the result of acquisition of second-line drug resistance in Nepal. Resistance to fluoroquinolones was detected among a high proportion of the Nepal isolates. This has implications for the management of TB, including appropriate antimicrobial stewardship and susceptibility testing for fluoroquinolones and other second-line TB drugs, to minimise the development of XDR-TB among Nepal TB cases.

Breed-specific reference sequence optimized mapping accuracy of NGS analyses for pigs

Background

Reference sequences play a vital role in next-generation sequencing (NGS), impacting mapping quality during genome analyses. However, reference genomes usually do not represent the full range of genetic diversity of a species as a result of geographical divergence and independent demographic events of different populations. For the mitochondrial genome (mitogenome), which occurs in high copy numbers in cells and is strictly maternally inherited, an optimal reference sequence has the potential to make mitogenome alignment both more accurate and more efficient. In this study, we used three different types of reference sequences for mitogenome mapping, i.e., the commonly used reference sequence (CU-ref), the breed-specific reference sequence (BS-ref) and the sample-specific reference sequence (SS-ref), respectively, and compared the accuracy of mitogenome alignment and SNP calling among them, for the purpose of proposing the optimal reference sequence for mitochondrial DNA (mtDNA) analyses of specific populations

Results

Four pigs, representing three different breeds, were high-throughput sequenced, subsequently mapping reads to the reference sequences mentioned above, resulting in a largest mapping ratio and a deepest coverage without increased running time when aligning reads to a BS-ref. Next, single nucleotide polymorphism (SNP) calling was carried out by 18 detection strategies with the three tools SAMtools, VarScan and GATK with different parameters, using the bam results mapping to BS-ref. The results showed that all eighteen strategies achieved the same high specificity and sensitivity, which suggested a high accuracy of mitogenome alignment by the BS-ref because of a low requirement for SNP calling tools and parameter choices.

Conclusions

This study showed that different reference sequences representing different genetic relationships to sample reads influenced mitogenome alignment, with the breed-specific reference sequences being optimal for mitogenome analyses, which provides a refined processing perspective for NGS data.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-021-08030-1.

Mycobacterium tuberculosis complex lineage 5 exhibits high levels of within-lineage genomic diversity and differing gene content compared to the type strain H37Rv

Pathogens of the Mycobacterium tuberculosis complex (MTBC) are considered to be monomorphic, with little gene content variation between strains. Nevertheless, several genotypic and phenotypic factors separate strains of the different MTBC lineages (L), especially L5 and L6 (traditionally termed Mycobacterium africanum) strains, from each other. However, this genome variability and gene content, especially of L5 strains, has not been fully explored and may be important for pathobiology and current approaches for genomic analysis of MTBC strains, including transmission studies. By comparing the genomes of 355 L5 clinical strains (including 3 complete genomes and 352 Illumina whole-genome sequenced isolates) to each other and to H37Rv, we identified multiple genes that were differentially present or absent between H37Rv and L5 strains. Additionally, considerable gene content variability was found across L5 strains, including a split in the L5.3 sub-lineage into L5.3.1 and L5.3.2. These gene content differences had a small knock-on effect on transmission cluster estimation, with clustering rates influenced by the selected reference genome, and with potential overestimation of recent transmission when using H37Rv as the reference genome. We conclude that full capture of the gene diversity, especially high-resolution outbreak analysis, requires a variation of the single H37Rv-centric reference genome mapping approach currently used in most whole-genome sequencing data analysis pipelines. Moreover, the high within-lineage gene content variability suggests that the pan-genome of M. tuberculosis is at least several kilobases larger than previously thought, implying that a concatenated or reference-free genome assembly (de novo) approach may be needed for particular questions.

One is not enough: On the effects of reference genome for the mapping and subsequent analyses of short-reads

Mapping of high-throughput sequencing (HTS) reads to a single arbitrary reference genome is a frequently used approach in microbial genomics. However, the choice of a reference may represent a source of errors that may affect subsequent analyses such as the detection of single nucleotide polymorphisms (SNPs) and phylogenetic inference. In this work, we evaluated the effect of reference choice on short-read sequence data from five clinically and epidemiologically relevant bacteria (Klebsiella pneumoniae, Legionella pneumophila, Neisseria gonorrhoeae, Pseudomonas aeruginosa and Serratia marcescens). Publicly available whole-genome assemblies encompassing the genomic diversity of these species were selected as reference sequences, and read alignment statistics, SNP calling, recombination rates, dN/dS ratios, and phylogenetic trees were evaluated depending on the mapping reference. The choice of different reference genomes proved to have an impact on almost all the parameters considered in the five species. In addition, these biases had potential epidemiological implications such as including/excluding isolates of particular clades and the estimation of genetic distances. These findings suggest that the single reference approach might introduce systematic errors during mapping that affect subsequent analyses, particularly for data sets with isolates from genetically diverse backgrounds. In any case, exploring the effects of different references on the final conclusions is highly recommended.

Mapping consists in the alignment of reads (i.e., DNA fragments) obtained through high-throughput genome sequencing to a previously assembled reference sequence. It is a common practice in genomic studies to use a single reference for mapping, usually the ‘reference genome’ of a species—a high-quality assembly. However, the selection of an optimal reference is hindered by intrinsic intra-species genetic variability, particularly in bacteria. It is known that genetic differences between the reference genome and the read sequences may produce incorrect alignments during mapping. Eventually, these errors could lead to misidentification of variants and biased reconstruction of phylogenetic trees (which reflect ancestry between different bacterial lineages). To our knowledge, this is the first work to systematically examine the effect of different references for mapping on the inference of tree topology as well as the impact on recombination and natural selection inferences. Furthermore, the novelty of this work relies on a procedure that guarantees that we are evaluating only the effect of the reference. This effect has proved to be pervasive in the five bacterial species that we have studied and, in some cases, alterations in phylogenetic trees could lead to incorrect epidemiological inferences. Hence, the use of different reference genomes may be prescriptive to assess the potential biases of mapping.

Is there a widespread clone of Serratia marcescens producing outbreaks worldwide?

BACKGROUND

Serratia marcescens frequently causes outbreaks in healthcare settings. There are few studies using high-throughput sequencing (HTS) that analyse S. marcescens outbreaks. We present the analysis of two outbreaks in neonatal intensive care units (NICUs) in hospitals from the Comunitat Valenciana (CV, Spain) and the impact of using different reference genomes.

METHODS

DNA from cultured isolates was extracted and sequenced by HTS using Illumina NextSeq. Reads were mapped against two reference genomes, strains UMH9 and Db11, and the unmapped fraction of the genomes was assembled to fully genetically characterize the isolates.

FINDINGS

Isolates from the first outbreak were identical to the UMH9 reference, an unrelated isolate obtained three years earlier in the USA. This did not occur when the Db11 strain, a standard reference for S. marcescens, was used as the reference for mapping. To check whether UMH9 was a widely distributed clone spreading in the CV, the second outbreak isolates were mapped against this reference. They were not closely related to this strain, and this outbreak could be defined as such regardless of the reference used for mapping the reads.

CONCLUSIONS

The choice of the reference for genomic analysis of outbreaks is a critical decision. In the case of the first outbreak, this choice changed the interpretation of the results drastically, allowing or preventing the definition of the outbreak according to the reference used. Although HTS is a powerful tool for epidemiological analysis, it is still essential to collect microbiological and epidemiological data for the correct interpretation of the results.

Genomic variant-identification methods may alter Mycobacterium tuberculosis transmission inferences

Pathogen genomic data are increasingly used to characterize global and local transmission patterns of important human pathogens and to inform public health interventions. Yet, there is no current consensus on how to measure genomic variation. To test the effect of the variant-identification approach on transmission inferences for Mycobacterium tuberculosis, we conducted an experiment in which five genomic epidemiology groups applied variant-identification pipelines to the same outbreak sequence data. We compared the variants identified by each group in addition to transmission and phylogenetic inferences made with each variant set. To measure the performance of commonly used variant-identification tools, we simulated an outbreak. We compared the performance of three mapping algorithms, five variant callers and two variant filters in recovering true outbreak variants. Finally, we investigated the effect of applying increasingly stringent filters on transmission inferences and phylogenies. We found that variant-calling approaches used by different groups do not recover consistent sets of variants, which can lead to conflicting transmission inferences. Further, performance in recovering true variation varied widely across approaches. While no single variant-identification approach outperforms others in both recovering true genome-wide and outbreak-level variation, variant-identification algorithms calibrated upon real sequence data or that incorporate local reassembly outperform others in recovering true pairwise differences between isolates. The choice of variant filters contributed to extensive differences across pipelines, and applying increasingly stringent filters rapidly eroded the accuracy of transmission inferences and quality of phylogenies reconstructed from outbreak variation. Commonly used approaches to identify M. tuberculosis genomic variation have variable performance, particularly when predicting potential transmission links from pairwise genetic distances. Phylogenetic reconstruction may be improved by less stringent variant filtering. Approaches that improve variant identification in repetitive, hypervariable regions, such as long-read assemblies, may improve transmission inference.

Mycobacterium bovis: From Genotyping to Genome Sequencing

Mycobacterium bovis is the main pathogen of bovine, zoonotic, and wildlife tuberculosis. Despite the existence of programs for bovine tuberculosis (bTB) control in many regions, the disease remains a challenge for the veterinary and public health sectors, especially in developing countries and in high-income nations with wildlife reservoirs. Current bTB control programs are mostly based on test-and-slaughter, movement restrictions, and post-mortem inspection measures. In certain settings, contact tracing and surveillance has benefited from M. bovis genotyping techniques. More recently, whole-genome sequencing (WGS) has become the preferential technique to inform outbreak response through contact tracing and source identification for many infectious diseases. As the cost per genome decreases, the application of WGS to bTB control programs is inevitable moving forward. However, there are technical challenges in data analyses and interpretation that hinder the implementation of M. bovis WGS as a molecular epidemiology tool. Therefore, the aim of this review is to describe M. bovis genotyping techniques and discuss current standards and challenges of the use of M. bovis WGS for transmission investigation, surveillance, and global lineages distribution. We compiled a series of associated research gaps to be explored with the ultimate goal of implementing M. bovis WGS in a standardized manner in bTB control programs.

Genomic diversity affects the accuracy of bacterial single-nucleotide polymorphism-calling pipelines

BACKGROUND

Accurately identifying single-nucleotide polymorphisms (SNPs) from bacterial sequencing data is an essential requirement for using genomics to track transmission and predict important phenotypes such as antimicrobial resistance. However, most previous performance evaluations of SNP calling have been restricted to eukaryotic (human) data. Additionally, bacterial SNP calling requires choosing an appropriate reference genome to align reads to, which, together with the bioinformatic pipeline, affects the accuracy and completeness of a set of SNP calls obtained. This study evaluates the performance of 209 SNP-calling pipelines using a combination of simulated data from 254 strains of 10 clinically common bacteria and real data from environmentally sourced and genomically diverse isolates within the genera Citrobacter, Enterobacter, Escherichia, and Klebsiella.

RESULTS

We evaluated the performance of 209 SNP-calling pipelines, aligning reads to genomes of the same or a divergent strain. Irrespective of pipeline, a principal determinant of reliable SNP calling was reference genome selection. Across multiple taxa, there was a strong inverse relationship between pipeline sensitivity and precision, and the Mash distance (a proxy for average nucleotide divergence) between reads and reference genome. The effect was especially pronounced for diverse, recombinogenic bacteria such as Escherichia coli but less dominant for clonal species such as Mycobacterium tuberculosis.

CONCLUSIONS

The accuracy of SNP calling for a given species is compromised by increasing intra-species diversity. When reads were aligned to the same genome from which they were sequenced, among the highest-performing pipelines was Novoalign/GATK. By contrast, when reads were aligned to particularly divergent genomes, the highest-performing pipelines often used the aligners NextGenMap or SMALT, and/or the variant callers LoFreq, mpileup, or Strelka.

Computational pan-genome mapping and pairwise SNP-distance improve detection of Mycobacterium tuberculosis transmission clusters

Next-generation sequencing based base-by-base distance measures have become an integral complement to epidemiological investigation of infectious disease outbreaks. This study introduces PANPASCO, a computational pan-genome mapping based, pairwise distance method that is highly sensitive to differences between cases, even when located in regions of lineage specific reference genomes. We show that our approach is superior to previously published methods in several datasets and across different Mycobacterium tuberculosis lineages, as its characteristics allow the comparison of a high number of diverse samples in one analysis—a scenario that becomes more and more likely with the increased usage of whole-genome sequencing in transmission surveillance.

Tuberculosis still is a threat to global health. It is essential to detect and interrupt transmissions to stop the spread of this infectious disease. With the rising use of next-generation sequencing methods, its application in the surveillance of Mycobacterium tuberculosis has become increasingly important in the last years. The main goal of molecular surveillance is the identification of patient-patient transmission and cluster detection. The mutation rate of M. tuberculosis is very low and stable. Therefore, many existing methods for comparative analysis of isolates provide inadequate results since their resolution is too limited. There is a need for a method that takes every detectable difference into account. We developed PANPASCO, a novel approach for comparing pairs of isolates using all genomic information available for each pair. We combine improved SNP-distance calculation with the use of a pan-genome incorporating more than 100 M. tuberculosis reference genomes representing lineages 1-4 for read mapping prior to variant detection. We thereby enable the collective analysis and comparison of similar and diverse isolates associated with different M. tuberculosis strains.

Mycobacterium avium in Community and Household Water, Suburban Philadelphia, Pennsylvania, USA, 2010-2012

Attention to environmental sources of Mycobacterium avium complex (MAC) infection is a vital component of disease prevention and control. We investigated MAC colonization of household plumbing in suburban Philadelphia, Pennsylvania, USA. We used variable-number tandem-repeat genotyping and whole-genome sequencing with core genome single-nucleotide variant analysis to compare M. avium from household plumbing biofilms with M. avium isolates from patient respiratory specimens. M. avium was recovered from 30 (81.1%) of 37 households, including 19 (90.5%) of 21 M. avium patient households. For 11 (52.4%) of 21 patients with M. avium disease, isolates recovered from their respiratory and household samples were of the same genotype. Within the same community, 18 (85.7%) of 21 M. avium respiratory isolates genotypically matched household plumbing isolates. Six predominant genotypes were recovered across multiple households and respiratory specimens. M. avium colonizing municipal water and household plumbing may be a substantial source of MAC pulmonary infection.

Genomic sequencing is required for identification of tuberculosis transmission in Hawaii

Background

Tuberculosis (TB) caused an estimated 1.4 million deaths and 10.4 million new cases globally in 2015. TB rates in the United States continue to steadily decline, yet rates in the State of Hawaii are perennially among the highest in the nation due to a continuous influx of immigrants from the Western Pacific and Asia. TB in Hawaii is composed of a unique distribution of genetic lineages, with the Beijing and Manila families of Mycobacterium tuberculosis (Mtb) comprising over two-thirds of TB cases. Standard fingerprinting methods (spoligotyping plus 24-loci Mycobacterial Interspersed Repetitive Units-Variable Number Tandem Repeats [MIRU-VNTR] fingerprinting) perform poorly when used to identify actual transmission clusters composed of isolates from these two families. Those typing methods typically group isolates from these families into large clusters of non-linked isolates with identical fingerprints. Next-generation whole-genome sequencing (WGS) provides a new tool for molecular epidemiology that can resolve clusters of isolates with identical spoligotyping and MIRU-VNTR fingerprints.

Methods

We performed WGS and SNP analysis and evaluated epidemiological data to investigate 19 apparent TB transmission clusters in Hawaii from 2003 to 2017 in order to assess WGS’ ability to resolve putative Mtb clusters from the Beijing and Manila families. This project additionally investigated MIRU-VNTR allele prevalence to determine why standard Mtb fingerprinting fails to usefully distinguish actual transmission clusters from these two Mtb families.

Results

WGS excluded transmission events in seven of these putative clusters, confirmed transmission in eight, and identified both transmission-linked and non-linked isolates in four. For epidemiologically identified clusters, while the sensitivity of MIRU-VNTR fingerprinting for identifying actual transmission clusters was found to be 100%, its specificity was only 28.6% relative to WGS. We identified that the Beijing and Manila families’ significantly lower Shannon evenness of MIRU-VNTR allele distributions than lineage 4 was the cause of standard fingerprinting’s poor performance when identifying transmission in Beijing and Manila family clusters.

Conclusions

This study demonstrated that WGS is necessary for epidemiological investigation of TB in Hawaii and the Pacific.

Electronic supplementary material

The online version of this article (10.1186/s12879-018-3502-1) contains supplementary material, which is available to authorized users.

Within-host Mycobacterium tuberculosis diversity and its utility for inferences of transmission

Whole genome sequencing in conjunction with traditional epidemiology has been used to reconstruct transmission networks of Mycobacterium tuberculosis during outbreaks. Given its low mutation rate, genetic diversity within M. tuberculosis outbreaks can be extremely limited - making it difficult to determine precisely who transmitted to whom. In addition to consensus SNPs (cSNPs), examining heterogeneous alleles (hSNPs) has been proposed to improve resolution. However, few studies have examined the potential biases in detecting these hSNPs. Here, we analysed genome sequence data from 25 specimens from British Columbia, Canada. Specimens were sequenced to a depth of 112-296×. We observed biases in read depth, base quality, strand distribution and read placement where possible hSNPs were initially identified, so we applied conservative filters to reduce false positives. Overall, there was phylogenetic concordance between the observed 2542 cSNP and 63 hSNP loci. Furthermore, we identified hSNPs shared exclusively by epidemiologically linked patients, supporting their use in transmission inferences. We conclude that hSNPs may add resolution to transmission networks, particularly where the overall genetic diversity is low.

Recombination Signal in Mycobacterium tuberculosis Stems from Reference-guided Assemblies and Alignment Artefacts

DNA acquisition via genetic recombination is considered advantageous as it has the potential to bring together beneficial mutations that emerge independently within a population. Furthermore, recombination is considered to contribute to the maintenance of genome stability by purging slightly deleterious mutations. The prevalence of recombination differs among prokaryotic species and depends on the accessibility of DNA transfer mechanisms. An exceptional example is the human pathogen Mycobacterium tuberculosis (MTB) where no clear transfer mechanisms have been so far characterized and the presence of recombination is questioned. Here, we analyze completely assembled MTB genomes in search for evidence of recombination. We find that putative recombination events are enriched in strains reconstructed by reference-guided assembly and in regions with unreliable alignments. In addition, assembly and alignment artefacts introduce phylogenetic signals that are conflicting the established MTB phylogeny. Our results reveal that the so far reported recombination events in MTB are likely to stem from methodological artefacts. We conclude that no reliable signal of recombination is observed in the currently available MTB genomes. Moreover, our study demonstrates the limitations of reference-guided genome assembly for phylogenetic reconstructions. Rigorously de novo assembled genomes of high quality are mandatory in order to distinguish true evolutionary signal from noise, in particular for low diversity species such as MTB.

Evaluating the use of whole-genome sequencing for outbreak investigations in the lack of closely related reference genome

Whole-genome sequencing (WGS) has emerged as a powerful molecular typing method for outbreak analysis enabling the rapid discrimination between outbreak and non-outbreak isolates. However, such analysis can be challenging in the absence of closely related reference genomes. In this study, we assessed the use of WGS in investigating an outbreak of a relatively understudied bacterial pathogen with no publicly available closely related reference genome. Eleven Burkholderia cepacia complex (Bcc) isolates (seven from patients and four from disposable dermal gloves packages) that were collected during an outbreak were sequenced using the Illumina MiSeq platform. Our results showed that mapping the 11 sequenced Bcc outbreak isolates against a genetically distant reference genome yield loses coverage (31.6-48.3%) and a high number of detected false single-nucleotide polymorphisms (SNPs) (1123-2139). Therefore, a reference genome consensus from an outbreak clinical isolate was generated by combining both de novo assembly and mapping approaches. Based on this approach, we were able to demonstrate that the Bcc outbreak isolates were closely related and were phylogenetically distinct from the 11 publically available Bcc genomes. In addition, the pairwise SNP distance analysis detected only 1 to 6 SNPs differences among the outbreak isolates, confirming that contaminated disposable dermal gloves were the cause of the outbreak.

A brief primer on genomic epidemiology: lessons learned from Mycobacterium tuberculosis

Genomics is now firmly established as a technique for the investigation and reconstruction of communicable disease outbreaks, with many genomic epidemiology studies focusing on revealing transmission routes of Mycobacterium tuberculosis. In this primer, we introduce the basic techniques underlying transmission inference from genomic data, using illustrative examples from M. tuberculosis and other pathogens routinely sequenced by public health agencies. We describe the laboratory and epidemiological scenarios under which genomics may or may not be used, provide an introduction to sequencing technologies and bioinformatics approaches to identifying transmission-informative variation and resistance-associated mutations, and discuss how variation must be considered in the light of available clinical and epidemiological information to infer transmission.