Abstract
Pulmonary tuberculosis (TB), caused by Mycobacterium tuberculosis, is a complex disease. The risk of developing active TB is in part determined by host genetic factors. Most genetic studies investigating TB susceptibility fail to replicate association signals particularly across diverse populations. South African populations arose because of multi-wave genetic admixture from the indigenous KhoeSan, Bantu-speaking Africans, Europeans, Southeast Asian-and East Asian populations. This has led to complex genetic admixture with heterogenous patterns of linkage disequilibrium and associated traits. As a result, precise estimation of both global and local ancestry is required to prevent both false positive and false-negative associations. Here, 820 individuals from South Africa were genotyped on the SNP-dense Illumina Multi-Ethnic Genotyping Array (∼1.7M SNPs) followed by local and global ancestry inference using RFMix. Local ancestry adjusted allelic association (LAAA) models were utilized owing to the extensive genetic heterogeneity present in this population. Hence, an interaction term, comprising the identification of the minor allele that corresponds to the ancestry present at the specific locus under investigation, was included as a covariate. One SNP (rs28647531) located on chromosome 4q22 was significantly associated with TB susceptibility and displayed a SNP minor allelic effect (G allele, frequency = 0.204) whilst correcting for local ancestry for Bantu-speaking African ancestry (p-value = 5.518 × 10−7; OR = 3.065; SE = 0.224). Although no other variants passed the significant threshold, clear differences were observed between the lead variants identified for each ancestry. Furthermore, the LAAA model robustly captured the source of association signals in multi-way admixed individuals from South Africa and allowed the identification of ancestry-specific disease risk alleles associated with TB susceptibility that have previously been missed.
Highlights
Pulmonary tuberculosis (TB), caused by the bacillus Mycobacterium tuberculosis (M.tb), is a complex disease which affects populations disproportionately and results from a multifactorial interaction between host and pathogen (Yim and Selvaraj, 2010)
Global ancestry deconvolution suggested a five-way admixed scenario for the study cohort. This is in accordance with previous studies. This diverse admixture and associated regional heterogeneity are reflected in the karyograms generated via local ancestry inference (Figure 2)
This scale of genetic heterogeneity suggests that no two individuals will harbour the same DNA segment from the same ancestral population, i.e., there is a high degree of locus-specific ancestry (Duan et al, 2018)
Summary
Pulmonary tuberculosis (TB), caused by the bacillus Mycobacterium tuberculosis (M.tb), is a complex disease which affects populations disproportionately and results from a multifactorial interaction between host and pathogen (Yim and Selvaraj, 2010). According to the World Health Organization (WHO), an estimated 10 million TB cases and 1.5 million deaths were reported in 2019 (WHO, 2019). TB remains a global health burden and is of particular concern in low- to middleincome countries where a generally higher incidence rate (615 per 100 000 in South Africa) occurs, together with the limitations of currently available therapies and vaccines (Bao et al, 2016; WHO | Global tuberculosis report 2019, 2019). The variation observed between populations from diverse geographic regions indicates possible ancestry-specific differences that contribute to the host genetic variability observed in TB genome-wide association studies (GWAS) (van Helden et al, 2006; Chimusa et al, 2014; Schurz et al, 2019a; Cai et al, 2019)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
