Liliana Kokusanilwa Rutaihwa

Genome-wide association studies rely on the statistical inference of untyped variants, called imputation, to increase the coverage of genotyping arrays. However, the results are often suboptimal in populations underrepresented in existing reference panels and array designs, since the selected single nucleotide polymorphisms (SNPs) may fail to capture population-specific haplotype structures, hence the full extent of common genetic variation. Here, we propose to sequence the full genomes of a small subset of an underrepresented study cohort to inform the selection of population-specific add-on tag SNPs and to generate an internal population-specific imputation reference panel, such that the remaining array-genotyped cohort could be more accurately imputed. Using a Tanzania-based cohort as a proof-of-concept, we demonstrate the validity of our approach by showing improvements in imputation accuracy after the addition of our designed add-on tags to the base H3Africa array.

Lineage 1 (L1) and 3 (L3) are two lineages of the Mycobacterium tuberculosis complex (MTBC), causing tuberculosis (TB) in humans. L1 and L3 are endemic to the Rim of the Indian Ocean, the region that accounts for most of the world’s new TB cases. Despite their relevance for this region, L1 and L3 remain understudied. Here we analyzed 2,938 L1 and 2,030 L3 whole genome sequences originating from 69 countries. We show that South Asia played a central role in the dispersion of these two lineages to neighboring regions. Moreover, we found that L1 exhibits signatures of local adaptation at the esxH locus, a gene coding for a secreted effector that targets the human endosomal sorting complex, and is included in several vaccine candidates. Our study highlights the importance of genetic diversity in the MTBC, and sheds new light on two of the most important MTBC lineages affecting humans.

The Lineage 2–Beijing (L2–Beijing) sub-lineage of Mycobacterium tuberculosis has received much attention due to its high virulence, fast disease progression, and association with antibiotic resistance. Despite several reports of the recent emergence of L2–Beijing in Africa, no study has investigated the evolutionary history of this sub-lineage on the continent. In this study, we used whole genome sequences of 781 L2 clinical strains from 14 geographical regions globally distributed to investigate the origins and onward spread of this lineage in Africa. Our results reveal multiple introductions of L2–Beijing into Africa linked to independent bacterial populations from East- and Southeast Asia. Bayesian analyses further indicate that these introductions occurred during the past 300 years, with most of these events pre-dating the antibiotic era. Hence, the success of L2–Beijing in Africa is most likely due to its hypervirulence and high transmissibility rather than drug resistance.

Human tuberculosis (TB) is caused by seven phylogenetic lineages of the Mycobacterium tuberculosis complex (MTBC), Lineage 1–7. Recent advances in rapid genotyping of MTBC based on single nucleotide polymorphisms (SNP), allow for phylogenetically robust strain classification, paving the way for defining genotype-phenotype relationships in clinical settings. Such studies have revealed that, in addition to host and environmental factors, strain variation in the MTBC influences the outcome of TB infection and disease. In Tanzania, such molecular epidemiological studies of TB however are scarce in spite of a high TB burden. Here we used SNP-typing to characterize a nationwide collection of 2,039 MTBC clinical isolates representative of 1.6% of all new and retreatment TB cases notified in Tanzania during 2012 and 2013. Four lineages, namely Lineage 1–4 were identified within the study population. The distribution and frequency of these lineages varied across regions but overall, Lineage 4 was the most frequent (n = 866, 42.5%), followed by Lineage 3 (n = 681, 33.4%) and 1 (n = 336, 16.5%), with Lineage 2 being the least frequent (n = 92, 4.5%). We found Lineage 2 to be independently associated with female sex (adjusted odds ratio [aOR] 2.14; 95% confidence interval [95% CI] 1.31 – 3.50, p = 0.002) and retreatment cases (aOR 1.67; 95% CI 0.95 – 2.84, p = 0. 065) in the study population. We found no associations between MTBC lineage and patient age or HIV status. Our sublineage typing based on spacer oligotyping on a subset of Lineage 1, 3 and 4 strains revealed the presence of mainly EAI, CAS and LAM families. Finally, we detected low levels of multidrug resistant isolates among a subset of 144 retreatment cases. This study provides novel insights into the MTBC lineages and the possible influence of pathogen–related factors on the TB epidemic in Tanzania.

Large sequence datasets are difficult to visualize and handle. Additionally, they often do not represent a random subset of the natural diversity, but the result of uncoordinated and convenience sampling. Consequently, they can suffer from redundancy and sampling biases. Here we present Treemmer, a simple tool to evaluate the redundancy of phylogenetic trees and reduce their complexity by eliminating leaves that contribute the least to the tree diversity. Treemmer can reduce the size of datasets with different phylogenetic structures and levels of redundancy while maintaining a sub-sample that is representative of the original diversity. Additionally, it is possible to fine-tune the behavior of Treemmer including any kind of meta-information, making Treemmer particularly useful for empirical studies.

Differences in rural and urban settings could account for distinct characteristics in the epidemiology of tuberculosis (TB). We comparatively studied epidemiological features of TB and helminth co-infections in adult patients from rural and urban settings of Tanzania. Adult patients (≥ 18 years) with microbiologically confirmed pulmonary TB were consecutively enrolled into two cohorts in Dar es Salaam, with ~ 4.4 million inhabitants (urban), and Ifakara in the sparsely populated Kilombero District with ~ 400 000 inhabitants (rural). Clinical data were obtained at recruitment. Stool and urine samples were subjected to diagnose helminthiases using Kato-Katz, Baermann, urine filtration, and circulating cathodic antigen tests. Differences between groups were assessed by χ2, Fisher’s exact, and Wilcoxon rank sum tests. Logistic regression models were used to determine associations. Between August 2015 and February 2017, 668 patients were enrolled, 460 (68.9%) at the urban and 208 (31.1%) at the rural site. Median patient age was 35 years (interquartile range [IQR]: 27–41.5 years), and 454 (68%) were males. Patients from the rural setting were older (median age 37 years vs. 34 years, P = 0.003), had a lower median body mass index (17.5 kg/m2 vs. 18.5 kg/m2, P < 0.001), a higher proportion of recurrent TB cases (9% vs. 1%, P < 0.001), and in HIV/TB co-infected patients a lower median CD4 cell counts (147 cells/μl vs. 249 cells/μl, P = 0.02) compared to those from urban Tanzania. There was no significant difference in frequencies of HIV infection, diabetes mellitus, and haemoglobin concentration levels between the two settings. The overall prevalence of helminth co-infections was 22.9% (95% confidence interval [CI]: 20.4–27.0%). The significantly higher prevalence of helminth infections at the urban site (25.7% vs. 17.3%, P = 0.018) was predominantly driven by Strongyloides stercoralis (17.0% vs. 4.8%, P < 0.001) and Schistosoma mansoni infection (4.1% vs. 16.4%, P < 0.001). Recurrent TB was associated with living in a rural setting (adjusted odds ratio [aOR]: 3.97, 95% CI: 1.16–13.67) and increasing age (aOR: 1.06, 95% CI: 1.02–1.10). Clinical characteristics and helminth co-infections pattern differ in TB patients in urban and rural Tanzania. The differences underline the need for setting-specific, tailored public health interventions to improve clinical management of TB and comorbidities.

Sebastien Gagneux and colleagues analyze a global collection of Mycobacterium tuberculosis clinical isolates to classify sublineages by phylogeography. They find globally distributed ‘generalist’ and geographically restricted ‘specialist’ sublineages of lineage 4, indicating that different evolutionary strategies were adopted to succeed in various ecological niches. Generalist and specialist species differ in the breadth of their ecological niches. Little is known about the niche width of obligate human pathogens. Here we analyzed a global collection of Mycobacterium tuberculosis lineage 4 clinical isolates, the most geographically widespread cause of human tuberculosis. We show that lineage 4 comprises globally distributed and geographically restricted sublineages, suggesting a distinction between generalists and specialists. Population genomic analyses showed that, whereas the majority of human T cell epitopes were conserved in all sublineages, the proportion of variable epitopes was higher in generalists. Our data further support a European origin for the most common generalist sublineage. Hence, the global success of lineage 4 reflects distinct strategies adopted by different sublineages and the influence of human migration.