A conformal Bayesian network for classification of Mycobacterium tuberculosis complex lineages

Genetic diversity of Mycobacterium tuberculosis strains isolated from spiritual holy water site attendees in Northwest Ethiopia. A cross-sectional study

Background

The genetic diversity of Mycobacterium tuberculosis complex (MTBC) strains was characterized among isolates from individuals with pulmonary tuberculosis (PTB) symptoms attended holy water sites (HWSs) in the Amhara region, Ethiopia.

Methods

A cross-sectional study was done from June 2019 to March 2020 to describe the genetic diversity and drug-resistance profiles of MTBC isolates. Sputum specimens were collected and cultured in the Löwenstein-Jensen culture medium. Line Probe Assay, MTBDRplus VER 2.0, and MTBDRsl VER 2.0 were used to detect first-and second-line anti-TB drug-resistance patterns. A spoligotyping technique was utilized to characterize the genetic diversity. Statistical analysis was performed using STATA 15.

Results

Of 560 PTB-symptomatic participants, 122 (21.8%) were culture-positive cases. Spoligotyping of 116 isolates revealed diverse MTBC sublineages, with four major lineages: Euro-American (EA) (Lineage 4), East-African-Indian (EAI) (Lineage 3), Ethiopian (ETH) (Lineage 7), East Asian (EA) (Lineage 2). The majority (96.6%) of the isolates were EA (lineage 4) and EAI, with proportions of 54.3% and 42.2%, respectively. A total of 31 spoligotype patterns were identified, 26 of which were documented in the SITVIT2 database. Of these, there were 15 unique spoligotypes, while eleven were grouped with 2-17 isolates. SIT149/T3-ETH (n = 17), SIT26/CAS1-DELHI (n = 16), SIT25/CAS1-DELHI (n = 12), and SIT52/T2 (n = 11) spoligotypes were predominant. A rare spoligotype pattern: SIT41/Turkey and SIT1/Beijing, has also been identified in North Shewa. The overall clustering rate of sub-lineages with known SIT was 76.4%.Of the 122 culture-positive isolates tested, 16.4% were resistant to rifampicin (RIF) and/or isoniazid (INH). Multidrug-resistant TB (MDR-TB) was detected in 12.3% of isolates, five of which were fluoroquinolones (FLQs) resistant. SIT149/T3-ETH and SIT21/CAS1-KILI sublineages showed a higher proportion of drug resistance.

Conclusions

Diverse MTBC spoligotypes were identified, with the T and CAS families and EA (lineage 4) predominating. A high prevalence of drug-resistant TB, with SIT149/T3-ETH and CAS1-KILI sublineages comprising a greater share, was observed. A study with large sample size and a sequencing method with stronger discriminatory power is warranted to understand better the genetic diversity of circulating MTBC in this cohort of study, which would help to adopt targeted interventions.

Genetic diversity of Mycobacterium tuberculosis isolates from the central, eastern and southeastern Ethiopia

Introduction

The population structure of Mycobacterium tuberculosis complex (MTBC) in Ethiopia is diverse but dominated by Euro-American (Lineage 4) and East-African-Indian (Lineage 3) lineages. The objective of this study was to describe the genetic diversity of MTBC isolates in Central, Eastern and Southeastern Ethiopia.

Methods

A total of 223 MTBC culture isolates obtained from patients referred to Adama and Harar TB reference laboratories were spoligotyped. Demographic and clinical characteristics were collected.

Results

Six major lineages: Euro-American (Lineage 4), East-African-Indian (Lineage 3), East Asian (Lineage 2), Indo-Oceanic (Lineage 1), Mycobacterium africanum (Lineage 5 and Lineage 6) and Ethiopian (Lineage 7) were identified. The majority (94.6 %) of the isolates were Euro-American and East-African-Indian, with proportions of 75.3 % and 19.3 %, respectively. Overall, 77 different spoligotype patterns were identified of which 42 were registered in the SITVIT2 database. Of these, 27 spoligotypes were unique, while 15 were clustered with 2-49 isolates. SIT149/T3_ETH (n = 49), SIT53/T1 (n = 33), SIT21/CAS1_Kili (n = 24) and SIT41/Turkey (n = 11) were the dominant spoligotypes. A rare Beijing spoligotype pattern, SIT541, has also been identified in Eastern Ethiopia. The overall clustering rate of sub-lineages with known SIT was 71.3 %. Age group (25-34) was significantly associated with clustering.

Conclusion

We found a heterogeneous population structure of MTBC dominated by T and CAS families, and the Euro-American lineage. The identification of the Beijing strain, particularly the rare SIT541 spoligotype in Eastern Ethiopia, warrants a heightened surveillance plan, as little is known about this genotype. A large-scale investigation utilizing a tool with superior discriminatory power, such as whole genome sequencing, is necessary to gain a thorough understanding of the genetic diversity of MTBC in the nation, which would help direct the overall control efforts.

Drug Resistance Conferring Mutation and Genetic Diversity of Mycobacterium tuberculosis Isolates in Tuberculosis Lymphadenitis Patients; Ethiopia

Background

Tuberculosis lymphadenitis (TBLN) is a growing public health concern in Ethiopia. However, there is limited information available on gene mutations conferring drug resistance and genetic diversity of M. tuberculosis isolates from TBLN patients.

Methods

Drug resistance and genetic diversity analysis were done on 91 M. tuberculosis isolates from culture positive TBLN patients collected between 2016 and 2017. Detection of mutations conferring resistance was carried out using GenoType MTBDRplus VER 2.0. Thereafter, isolates were typed using spoligotyping.

Results

Out of the 91 strains, mutations conferring resistance to rifampicin (RIF) and isoniazid (INH) were observed in two (2.2%) and six (6.6%) isolates, respectively. The two RIF resistant isolates displayed a mutation at codon 531 in the rpoB gene with amino acid change of S531L. Among the six INH resistant strains, four isolates had shown mutation at the KatG gene at codon 315 with amino acid change of S315T, one isolate had a mutation at the inhA gene at codon 15 with amino acid change of C15T and one isolate had a mutation at the inhA gene with unknown amino acid change. All drug resistant isolates were from treatment naive TBLN patients. The dominantly identified Spoligo International Types (SITs) were SIT25, SIT149, and SIT53, respectively; these accounted for 43% of the total number of strains. The isolates were grouped into four main lineages; Lineage 1 (2, 2.2%), Lineage 3 (38, 41.7%), Lineage 4 (49, 53.8%) and Lineage 7 (2, 2.2%). Four out of six (66.7%) isolates with drug resistance conferring mutations belonged to clustered strains (strains with shared SIT).

Conclusion

The detection of drug resistant conferring mutation in treatment naïve TBLN patients together with detection of drug resistant isolates among clustered strains might suggest resistant strains' transmission in the community. This needs to be carefully considered to prevent the spread of drug resistant clones in the country.

Molecular Analysis of Mycobacterium tuberculosis Isolated in the North Central Zone of Nigeria

Tuberculosis (TB) incidence in Nigeria is high, with a significant burden of TB/Human Immunodeficiency Virus (HIV). Genotyping and drug susceptibility of Mycobacterium tuberculosis Complex (MTBC) are important in order to improve the control of the disease. This study sought to determine drug susceptibility and genetic diversity of MTBC in the country. The sputum samples of 202 patients [133 (65.8%) males/69 (34.2%) females] were collected in the North Central zone of Nigeria and cultured using Lowenstein–Jensen medium. Immunochromatography for the primary identification and Drug Susceptibility Testing (DST) by proportion method, as well as IS6110 typing, regions of difference 1, 4, 9, 12, 702, and 711, and spoligotyping were carried out on the isolates. Following the DST on 202 isolates, 51 (25.2%) showed resistance to at least one drug. Multidrug resistance was observed in 29/202 (14.4%) cases. HIV positivity [37/202 (18.3%) patients] was associated with rifampicin 9/37 (24.3%) resistance (p = 0.012) as well as gender (p = 0.009). Of the 202 isolates, 150 (74.3%) were identified as the Cameroon sublineage, followed by the UgandaI, Haarlem, and West Africa 1 with 18 (8.9%), 10 (5%), and 6 (3%), respectively. The LAM10_CAM was the most prevalent genetic family [128/202 (63.4%)], with the shared international type 61 [111 (55%) isolates] the largest cluster. Gender (p = 0.038) and age (p = 0.015) had significant associations with the LAM10_CAM family but neither with HIV (p = 0.479) nor drug resistance. Rifampicin resistance in TB/HIV coinfected patient is a major concern in the study area. The Mycobacterium africanum lineage showed a marked decrease, and the need to educate females most at risk of TB/HIV coinfection is advocated.

Resolving a clinical tuberculosis outbreak using palaeogenomic genome reconstruction methodologies

This study describes the analysis of DNA from heat-killed (boilate) isolates of Mycobacterium tuberculosis from two UK outbreaks where DNA was of sub-optimal quality for the standard methodologies routinely used in microbial genomics. An Illumina library construction method developed for sequencing ancient DNA was successfully used to obtain whole genome sequences, allowing analysis of the outbreak by gene-by-gene MLST, SNP mapping and phylogenetic analysis. All cases were spoligotyped to the same Haarlem H1 sub-lineage. This is the first described application of ancient DNA library construction protocols to allow whole genome sequencing of a clinical tuberculosis outbreak. Using this method it is possible to obtain epidemiologically meaningful data even when DNA is of insufficient quality for standard methods.

Global variation in bacterial strains that cause tuberculosis disease: a systematic review and meta-analysis

BACKGROUND

The host, microbial, and environmental factors that contribute to variation in tuberculosis (TB) disease are incompletely understood. Accumulating evidence suggests that one driver of geographic variation in TB disease is the local ecology of mycobacterial genotypes or strains, and there is a need for a comprehensive and systematic synthesis of these data. The objectives of this study were to (1) map the global distribution of genotypes that cause TB disease and (2) examine whether any epidemiologically relevant clinical characteristics were associated with those genotypes.

METHODS

We performed a systematic review of PubMed and Scopus to create a comprehensive dataset of human TB molecular epidemiology studies that used representative sampling techniques. The methods were developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). We extracted and synthesized data from studies that reported prevalence of bacterial genotypes and from studies that reported clinical characteristics associated with those genotypes.

RESULTS

The results of this study are twofold. First, we identified 206 studies for inclusion in the study, representing over 200,000 bacterial isolates collected over 27 years in 85 countries. We mapped the genotypes and found that, consistent with previously published maps, Euro-American lineage 4 and East Asian lineage 2 strains are widespread, and West African lineages 5 and 6 strains are geographically restricted. Second, 30 studies also reported transmission chains and 4 reported treatment failure associated with genotypes. We performed a meta-analysis and found substantial heterogeneity across studies. However, based on the data available, we found that lineage 2 strains may be associated with increased risk of transmission chains, while lineages 5 and 6 strains may be associated with reduced risk, compared with lineage 4 strains.

CONCLUSIONS

This study provides the most comprehensive systematic analysis of the evidence for diversity in bacterial strains that cause TB disease. The results show both geographic and epidemiological differences between strains, which could inform our understanding of the global burden of TB. Our findings also highlight the challenges of collecting the clinical data required to inform TB diagnosis and treatment. We urge future national TB programs and research efforts to prioritize and reinforce clinical data collection in study designs and results dissemination.

Molecular characterization of Mycobacterium tuberculosis isolates from Tanga, Tanzania: First insight of MIRU-VNTR and microarray-based spoligotyping in a high burden country

Molecular typing of Mycobacterium tuberculosis(MTB) has greatly enhanced the understanding of the population structure of MTB isolates and epidemiology of tuberculosis (TB). To characterize prevalent genotypes of MTB, microarrays‑based spoligotyping and mycobacterial interspersed repetitive unit‑variable number of tandem repeats (MIRU‑VNTR) were applied on 80 isolates collected from primary health care facilities in Tanga, North‑eastern Tanzania. A total of 18 distinct spoligotypes were identified. The lineages by order of their predominance were EAI and CAS families (26.25%, 21 isolates each), LAM family and T super‑family (10%, 8 isolates each), MANU family (3.75%, 3 isolates), Beijing family (2.5%, 2 isolates) and S family (1.25%, 1 isolate). Overall, sixteen (20%) strains could not be allocated to any lineage according to the SITVIT_WEB database. The allelic diversity (h) for specific MIRU‑VNTR loci showed a considerable variation ranging from 0.826 of VNTR locus 3192 to 0.141 of VNTR locus 2059. The allelic diversity for 11 loci (VNTR 3192, 2996, 2165, 960, 4052, 424, 4156, 2531, 1644, 802 and 3690) exceeded 0.6, indicating highly discriminatory power. Seven loci (VNTR 2163b, 2401, 1955, 577, 4348, 2687 and 580) showed moderate discrimination (0.3 ≤ h ≥ 0.6), and three loci (VNTR3007, 154 and 2059) were less polymorphic. The present study suggests that the TB cases in Tanga might be caused by a diverse array of MTB strain families that may be indicative of a cosmopolitan population with frequent migration and travel. Microarray‑based spoligotyping and MIRU‑VNTR could be reliable tools in detecting different MTB genotypes in high burden settings.

A Mycobacterial Perspective on Tuberculosis in West Africa: Significant Geographical Variation of M. africanum and Other M. tuberculosis Complex Lineages

BACKGROUND

Phylogenetically distinct Mycobacterium tuberculosis lineages differ in their phenotypes and pathogenicity. Consequently, understanding mycobacterial population structures phylogeographically is essential for design, interpretation and generalizability of clinical trials. Comprehensive efforts are lacking to date to establish the West African mycobacterial population structure on a sub-continental scale, which has diagnostic implications and can inform the design of clinical TB trials.

METHODOLOGY/PRINCIPAL FINDINGS

We collated novel and published genotyping (spoligotyping) data and classified spoligotypes into mycobacterial lineages/families using TBLineage and Spotclust, followed by phylogeographic analyses using statistics (logistic regression) and lineage axis plot analysis in GenGIS, in which a phylogenetic tree constructed in MIRU-VNTRplus was analysed. Combining spoligotyping data from 16 previously published studies with novel data from The Gambia, we obtained a total of 3580 isolates from 12 countries and identified 6 lineages comprising 32 families. By using stringent analytical tools we demonstrate for the first time a significant phylogeographic separation between western and eastern West Africa not only of the two M. africanum (West Africa 1 and 2) but also of several major M. tuberculosis sensu stricto families, such as LAM10 and Haarlem 3. Moreover, in a longitudinal logistic regression analysis for grouped data we showed that M. africanum West Africa 2 remains a persistent health concern.

CONCLUSIONS/SIGNIFICANCE

Because of the geographical divide of the mycobacterial populations in West Africa, individual research findings from one country cannot be generalized across the whole region. The unequal geographical family distribution should be considered in placement and design of future clinical trials in West Africa.

Predicting Mycobacterium tuberculosis complex clades using knowledge-based Bayesian networks

We develop a novel approach for incorporating expert rules into Bayesian networks for classification of Mycobacterium tuberculosis complex (MTBC) clades. The proposed knowledge-based Bayesian network (KBBN) treats sets of expert rules as prior distributions on the classes. Unlike prior knowledge-based support vector machine approaches which require rules expressed as polyhedral sets, KBBN directly incorporates the rules without any modification. KBBN uses data to refine rule-based classifiers when the rule set is incomplete or ambiguous. We develop a predictive KBBN model for 69 MTBC clades found in the SITVIT international collection. We validate the approach using two testbeds that model knowledge of the MTBC obtained from two different experts and large DNA fingerprint databases to predict MTBC genetic clades and sublineages. These models represent strains of MTBC using high-throughput biomarkers called spacer oligonucleotide types (spoligotypes), since these are routinely gathered from MTBC isolates of tuberculosis (TB) patients. Results show that incorporating rules into problems can drastically increase classification accuracy if data alone are insufficient. The SITVIT KBBN is publicly available for use on the World Wide Web.

Molecular epidemiology and genotyping of Mycobacterium tuberculosis isolated in Baghdad

Tuberculosis (TB) remains a major health problem in Iraq but the strains responsible for the epidemic have been poorly characterized. Our aim was to characterize the TB strains circulating in Bagdad (Iraq). A total of 270 Mycobacterium tuberculosis complex (MTBC) strains isolated between 2010 and 2011 from TB patients attending the Center of Chest and Respiratory diseases in Baghdad were analyzed by Spoligotyping. The analysis indicated that 94.1% of the isolates belong to known genotype clades: CAS 39.6%, ill-defined T clade 29.6%, Manu 7.4%, Haarlem 7%, Ural 4.1%, LAM 3.3%, X 0.7%, LAM7-TUR 0.7%, EAI 0.7%, S 0.7%, and unknown 5.9%. Comparison with the international multimarker database SITVIT2 showed that SIT 309 (CAS1-Delhi) and SIT1144 (T1) were the most common types. In addition, 44 strains were included in SITVIT2 database under 16 new Spoligotype International Types (SITs); of these, 6 SITs (SIT3346, SIT3497, SIT3708, SIT3790, SIT3791, and SIT3800) (n = 32 strains) were created within the present study and 10 were created after a match with an orphan in the database. By using 24-loci MIRU-VNTR-typing on a subset of 110 samples we found a high recent transmission index (RTI) of 33.6%. In conclusion, we present the first unifying framework for both epidemiology and evolutionary analysis of M. tuberculosis in Iraq.

Genotype diversity of Mycobacterium isolates from children in Jimma, Ethiopia

BACKGROUND

Paediatric tuberculosis (TB) is poorly addressed in Ethiopia and information about its magnitude and the genotype distribution of the causative Mycobacterium tuberculosis strains responsible for its spread are scanty.

METHODS

Gastric lavage or sputum samples were collected from consecutively enrolled TB suspect children visiting Jimma University Hospital in 2011 and cultured on Middlebrook 7H11 and Löwenstein-Jensen media. Acid fast bacterial (AFB) isolates were subjected to molecular typing targeting regions of difference (RDs), 16S rDNA gene and the direct repeat (DR) region using multiplex polymerase chain reaction (mPCR), gene sequencing and spoligotyping, respectively. Molecular drug susceptibility testing of M. tuberculosis isolates was performed by Genotype®MTBDRplus line probe assay (LPA) (Hain Life Sciences, Germany).

RESULTS

Gastric lavage (n = 43) or sputum (n = 58) samples were collected from 101 children and 31.7% (32/101) of the samples were positive for AFB by microscopy, culture and/or PCR. Out of 25 AFB isolates, 60% (15/25) were identified as M. tuberculosis by PCR, and 40% isolates (10/25) were confirmed to be non-tuberculous mycobacteria (NTM) by genus typing and 16S rDNA gene sequencing. Lineage classification assigned the M. tuberculosis strains into Euro-American (EUA, 66.7%; 10/15), East-African-Indian (EAI; 2/15), East-Asian (EA; 1/15) and Indio-Oceanic (IO; 1/15) lineages. Seven M. tuberculosis strains were new to the SpolDB4 database. All of the M. tuberculosis isolates were susceptible to isoniazid (INH) and rifampicin (RIF), except for one strain (of spoligotype SIT-149 or T3_ETH family) which had a mutation at the inhA locus which often confers resistance to INH (low level) and ethionamide.

CONCLUSIONS

Analysis of the genetic population structure of paediatric M. tuberculosis strains suggested similarity with that of adults, indicating an on-going and active transmission of M. tuberculosis from adults to children in Ethiopia. There were no multidrug-resistant TB (MDR-TB) strains among the isolates.

Genotypes of Mycobacterium tuberculosis in patients at risk of drug resistance in Bolivia

Bolivia ranks among the 10 Latin American countries with the highest rates of tuberculosis (TB) and multidrug resistant (MDR) TB. In view of this, and of the lacking information on the population structure of Mycobacterium tuberculosis in the country, we explored genotype associations with drug resistance and clustering by analyzing isolates collected in 2010 from 100 consecutive TB patients at risk of drug resistance in seven of the nine departments in which Bolivia is divided. Fourteen isolates were MDR, 29 had other drug resistance profiles, and 57 were pansusceptible. Spoligotype family distribution was: Haarlem 39.4%, LAM 26.3%, T 22.2%, S 2.0%, X 1.0%, orphan 9.1%, with very low intra-family diversity and absence of Beijing genotypes. We found 66 different MIRU-VNTR patterns; the most frequent corresponded to Multiple Locus Variable Analysis (MLVA) MtbC15 patterns 860, 372 and 873. Twelve clusters, each with identical MIRU-VNTR and spoligotypes, gathered 35 patients. We found no association of genotype with drug resistant or MDR-TB. Clustering associated with SIT 50 and the H3 subfamily to which it belongs (p<0.0001). The largest cluster involved isolates from three departments and displayed a genotype (SIT 50/MLVA 860) previously identified in Bolivian migrants into Spain and Argentina suggesting that this genotype is widespread among Bolivian patients. Our study presents a first overview of M. tuberculosis genotypes at risk of drug resistance circulating in Bolivia. However, results should be taken cautiously because the sample is small and includes a particular subset of M. tuberculosis population.

Current trends of Mycobacterium tuberculosis molecular epidemiology in Saudi Arabia and associated demographical factors

Data are scarce on demographical factors related to the population structure of Mycobacterium tuberculosis in Saudi Arabia. A study was conducted on 902 clinical isolates to explore current trends in the phylogeography and associated demographical factors of tuberculosis by using spoligotyping and 24 loci based MIRU-VNTR typing. Young male patients (aged 16-29 and 30-44) were predominant in this cohort. The phylogenetic diversity among M. tuberculosis isolates was found high, as almost all known genetic lineages were identified. Delhi/CAS (26.4%), EAI (13.7%) and Haarlem (11.3%) were the most common lineages observed, particularly among the low age groups (16-29 and 30-44 years), whereas elderly patients (>60 years) showed a predominance in the lineages S, Ghana, TUR and Uganda-I. A statistically significant association was observed between gender of the patients and lineages of EAI (p value 0.026) and LAM (p value 0.005). Overall, molecular strain cluster rate was 34.4% with an elevated rate among patients aged below 15 years (43.1%), while cases among the elderly (>60 years) showed the lowest degree of clustering (12.5%). The largest level of clustering was noticed among cases caused by strains of the lineages Haarlem (59.8%), Beijing (55.8%) and LAM (42.8%). The current population structure of M. tuberculosis in Saudi Arabia is highly diverse with significant associations to demography, transmission dynamics and origin of the patients. The difference in genotype distributions among low and high aged patients reflects the ongoing change in the strain population structure in the country.

Inferred spoligoforest topology unravels spatially bimodal distribution of mutations in the DR region

Biomarkers of Mycobacterium tuberculosis complex (MTBC) mutate over time. Among the biomarkers of MTBC, spacer oligonucleotide type (spoligotype) and mycobacterium interspersed repetitive unit (MIRU) patterns are commonly used to genotype clinical MTBC strains. In this study, we present an evolution model of spoligotype rearrangements using MIRU patterns to disambiguate the ancestors of spoligotypes. We use a large patient dataset from the United States Centers for Disease Control and Prevention (CDC) to generate this model. Based on the contiguous deletion assumption and rare observation of convergent evolution, we first generate the most parsimonious forest of spoligotypes, called a spoligoforest, using three genetic distance measures. An analysis of topological attributes of the spoligoforest and number of variations at the direct repeat (DR) locus of each strain reveals interesting properties of deletions in the DR region. First, we compare our mutation model to existing mutation models of spoligotypes and find that our mutation model produces as many within-lineage mutation events as other models, with slightly higher segregation accuracy. Second, based on our mutation model, the number of descendant spoligotypes follows a power law distribution. Third, contrary to prior studies, the power law distribution does not plausibly fit to the mutation length frequency. Moreover, we find that the total number of mutation events at consecutive spacers follows a spatially bimodal distribution. The two modes are spacers 13 and 40, which are hotspots for chromosomal rearrangements, and the change point is spacer 34, which is absent in most MTBC strains. Based on this observation, we built two alternative models for mutation length frequency: the Starting Point Model (SPM) and the Longest Block Model (LBM). Both models are plausibly good fits to the mutation length frequency distribution, as verified by the goodness-of-fit test. We also apply SPM and LBM to a dataset from Institut Pasteur de Guadeloupe and verify that these models hold for different strain datasets.

TB-Lineage: an online tool for classification and analysis of strains of Mycobacterium tuberculosis complex

This paper formulates a set of rules to classify genotypes of the Mycobacterium tuberculosis complex (MTBC) into major lineages using spoligotypes and MIRU-VNTR results. The rules synthesize prior literature that characterizes lineages by spacer deletions and variations in the number of repeats seen at locus MIRU24 (alias VNTR2687). A tool that efficiently and accurately implements this rule base is now freely available at http://tbinsight.cs.rpi.edu/run_tb_lineage.html. When MIRU24 data is not available, the system utilizes predictions made by a Naïve Bayes classifier based on spoligotype data. This website also provides a tool to generate spoligoforests in order to visualize the genetic diversity and relatedness of genotypes and their associated lineages. A detailed analysis of the application of these tools on a dataset collected by the CDC consisting of 3198 distinct spoligotypes and 5430 distinct MIRU-VNTR types from 37,066 clinical isolates is presented. The tools were also tested on four other independent datasets. The accuracy of automated classification using both spoligotypes and MIRU24 is >99%, and using spoligotypes alone is >95%. This online rule-based classification technique in conjunction with genotype visualization provides a practical tool that supports surveillance of TB transmission trends and molecular epidemiological studies.

Web tools for molecular epidemiology of tuberculosis

In this study we explore publicly available web tools designed to use molecular epidemiological data to extract information that can be employed for the effective tracking and control of tuberculosis (TB). The application of molecular methods for the epidemiology of TB complement traditional approaches used in public health. DNA fingerprinting methods are now routinely employed in TB surveillance programs and are primarily used to detect recent transmissions and in outbreak investigations. Here we present web tools that facilitate systematic analysis of Mycobacterium tuberculosis complex (MTBC) genotype information and provide a view of the genetic diversity in the MTBC population. These tools help answer questions about the characteristics of MTBC strains, such as their pathogenicity, virulence, immunogenicity, transmissibility, drug-resistance profiles and host-pathogen associativity. They provide an integrated platform for researchers to use molecular epidemiological data to address current challenges in the understanding of TB dynamics and the characteristics of MTBC.

Sublineage structure analysis of Mycobacterium tuberculosis complex strains using multiple-biomarker tensors

BACKGROUND

Strains of Mycobacterium tuberculosis complex (MTBC) can be classified into major lineages based on their genotype. Further subdivision of major lineages into sublineages requires multiple biomarkers along with methods to combine and analyze multiple sources of information in one unsupervised learning model. Typically, spacer oligonucleotide type (spoligotype) and mycobacterial interspersed repetitive units (MIRU) are used for TB genotyping and surveillance. Here, we examine the sublineage structure of MTBC strains with multiple biomarkers simultaneously, by employing a tensor clustering framework (TCF) on multiple-biomarker tensors.

RESULTS

Simultaneous analysis of the spoligotype and MIRU type of strains using TCF on multiple-biomarker tensors leads to coherent sublineages of major lineages with clear and distinctive spoligotype and MIRU signatures. Comparison of tensor sublineages with SpolDB4 families either supports tensor sublineages, or suggests subdivision or merging of SpolDB4 families. High prediction accuracy of major lineage classification with supervised tensor learning on multiple-biomarker tensors validates our unsupervised analysis of sublineages on multiple-biomarker tensors.

CONCLUSIONS

TCF on multiple-biomarker tensors achieves simultaneous analysis of multiple biomarkers and suggest a new putative sublineage structure for each major lineage. Analysis of multiple-biomarker tensors gives insight into the sublineage structure of MTBC at the genomic level.

Using affinity propagation for identifying subspecies among clonal organisms: lessons from M. tuberculosis

Background

Classification and naming is a key step in the analysis, understanding and adequate management of living organisms. However, where to set limits between groups can be puzzling especially in clonal organisms. Within the Mycobacterium tuberculosis complex (MTC), the etiological agent of tuberculosis (TB), experts have first identified several groups according to their pattern at repetitive sequences, especially at the CRISPR locus (spoligotyping), and to their epidemiological relevance. Most groups such as "Beijing" found good support when tested with other loci. However, other groups such as T family and T1 subfamily (belonging to the "Euro-American" lineage) correspond to non-monophyletic groups and still need to be refined. Here, we propose to use a method called Affinity Propagation that has been successfully used in image categorization to identify relevant patterns at the CRISPR locus in MTC.

Results

To adequately infer the relative divergence time between strains, we used a distance method inspired by the recent evolutionary model by Reyes et al. We first confirm that this method performs better than the Jaccard index commonly used to compare spoligotype patterns. Second, we document the support of each spoligotype family among the previous classification using affinity propagation on the international spoligotyping database SpolDB4. This allowed us to propose a consensus assignation for all SpolDB4 spoligotypes. Third, we propose new signatures to subclassify the T family.

Conclusion

Altogether, this study shows how the new clustering algorithm Affinity Propagation can help building or refining clonal organims classifications. It also describes well-supported families and subfamilies among M. tuberculosis complex, especially inside the modern "Euro-American" lineage.

The forest behind the tree: phylogenetic exploration of a dominant Mycobacterium tuberculosis strain lineage from a high tuberculosis burden country

BACKGROUND

Genotyping of Mycobacterium tuberculosis isolates is a powerful tool for epidemiological control of tuberculosis (TB) and phylogenetic exploration of the pathogen. Standardized PCR-based typing, based on 15 to 24 mycobacterial interspersed repetitive unit-variable number of tandem repeat (MIRU-VNTR) loci combined with spoligotyping, has been shown to have adequate resolution power for tracing TB transmission and to be useful for predicting diverse strain lineages in European settings. Its informative value needs to be tested in high TB-burden countries, where the use of genotyping is often complicated by dominance of geographically specific, genetically homogeneous strain lineages.

METHODOLOGY/PRINCIPAL FINDINGS

We tested this genotyping system for molecular epidemiological analysis of 369 M. tuberculosis isolates from 3 regions of Brazil, a high TB-burden country. Deligotyping, targeting 43 large sequence polymorphisms (LSPs), and the MIRU-VNTRplus identification database were used to assess phylogenetic predictions. High congruence between the different typing results consistently revealed the countrywide supremacy of the Latin-American-Mediterranean (LAM) lineage, comprised of three main branches. In addition to an already known RDRio branch, at least one other branch characterized by a phylogenetically informative LAM3 spoligo-signature seems to be globally distributed beyond Brazil. Nevertheless, by distinguishing 321 genotypes in this strain population, combined MIRU-VNTR typing and spoligotyping demonstrated the presence of multiple distinct clones. The use of 15 to 24 loci discriminated 21 to 25% more strains within the LAM lineage, compared to a restricted lineage-specific locus set suggested to be used after SNP analysis. Noteworthy, 23 of the 28 molecular clusters identified were exclusively composed of patient isolates from a same region, consistent with expected patterns of mostly local TB transmission.

CONCLUSIONS/SIGNIFICANCE

Standard MIRU-VNTR typing combined with spoligotyping can reveal epidemiologically meaningful clonal diversity behind a dominant M. tuberculosis strain lineage in a high TB-burden country and is useful to explore international phylogenetical ramifications.

Data-driven insights into deletions of Mycobacterium tuberculosis complex chromosomal DR region using spoligoforests

Biomarkers of Mycobacterium tuberculosis complex (MTBC) mutate over time. Among the biomarkers of MTBC, spacer oligonucleotide type (spoligotype) and Mycobacterium Interspersed Repetitive Unit (MIRU) patterns are commonly used to genotype clinical MTBC strains. In this study, we present an evolution model of spoligotype rearrangements using MIRU patterns to disambiguate the ancestors of spoligotypes, in a large patient dataset from the United States Centers for Disease Control and Prevention (CDC). Based on the contiguous deletion assumption and rare observation of convergent evolution, we first generate the most parsimonious forest of spoligotypes, called a spoligoforest, using three genetic distance measures. An analysis of topological attributes of the spoligoforest and number of variations at the direct repeat (DR) locus of each strain reveals interesting properties of deletions in the DR region. First, we compare our mutation model to existing mutation models of spoligotypes and find that our mutation model produces as many within-lineage mutation events as other models, with slightly higher segregation accuracy. Second, based on our mutation model, the number of descendant spoligotypes follows a power law distribution. Third, contrary to prior studies, the power law distribution does not plausibly fit to the mutation length frequency. Finally, the total number of mutation events at consecutive DR loci follows a bimodal distribution, which results in accumulation of shorter deletions in the DR region. The two modes are spacers 13 and 40, which are hotspots for chromosomal rearrangements. The change point in the bimodal distribution is spacer 34, which is absent in most MTBC strains. This bimodal separation results in accumulation of shorter deletions, which explains why a power law distribution is not a plausible fit to the mutation length frequency.

Examining the sublineage structure of Mycobacterium tuberculosis complex strains with multiple-biomarker tensors

Strains of the Mycobacterium tuberculosis complex (MTBC) can be classified into coherent lineages of similar traits based on their genotype. We present a tensor clustering framework to group MTBC strains into sublineages of the known major lineages based on two biomarkers: spacer oligonucleotide type (spoligotype) and mycobacterial interspersed repetitive units (MIRU). We represent genotype information of MTBC strains in a high-dimensional array in order to include information about spoligotype, MIRU, and their coexistence using multiple-biomarker tensors. We use multiway models to transform this multidimensional data about the MTBC strains into two-dimensional arrays and use the resulting score vectors in a stable partitive clustering algorithm to classify MTBC strains into sublineages. We validate clusterings using cluster stability and accuracy measures, and find stabilities of each cluster. Based on validated clustering results, we present a sublineage structure of MTBC strains and compare it to the sublineage structures of SpolDB4 and MIRU-VNTRplus.