Investigating Transmission Patterns of Mycobacterium abscessus Clinical Isolates from a Korean Cohort Using Whole Genome Sequencing and Multi-Caller Variant Calling Methods
Jae Jin Lee
Division of Genome and Health Big Data
Graduate School of Public ...
Investigating Transmission Patterns of Mycobacterium abscessus Clinical Isolates from a Korean Cohort Using Whole Genome Sequencing and Multi-Caller Variant Calling Methods
Jae Jin Lee
Division of Genome and Health Big Data
Graduate School of Public Health
Seoul National University
Background: The Mycobacterium abscessus complex (MABC) has become an emerging pathogen responsible for chronic, multidrug-resistant nontuberculous mycobacterial (NTM) pulmonary infections. Among its subspecies, M. a. subsp. abscessus and M. a. subsp. massiliense are the most clinically relevant, yet their population structures and transmission remain poorly understood—particularly within Asian healthcare settings such as South Korea.
Methods: Whole-genome sequencing (WGS) was performed on 302 Korean clinical isolates (199 M. a. subsp. abscessus, 103 M. a. subsp. abscessus subsp. massiliense), supplemented with 173 publicly available genomes from previous global studies. High-confidence single-nucleotide variants (SNVs) were identified using a multi-caller consensus approach integrating GATK HaplotypeCaller, FreeBayes, and Bcftools. Phylogenetic relationships were reconstructed with RAxML, and pairwise SNV distances were calculated to define genomic clusters (≤50 SNVs) suggestive of potential epidemiological linkage. Additionally, a bacterial genome-wide association study (bGWAS) was performed using pyseer with a linear mixed model (LMM) framework incorporating genetic similarity matrices and multidimensional scaling (MDS) covariates to identify variants associated with clustering.
Results: Phylogenomic analysis revealed 38 clusters among 275 M. a. subsp. abscessus isolates, 28 of which comprised exclusively Korean clinical strains. In contrast, M. a. subsp. massiliense (n = 200) displayed 23 clusters, but only two isolate pairs showed ≤50 SNVs, consistent with its high genomic diversity and predominantly environmental acquisition. The bGWAS of M. a. subsp. abscessus identified several loci significantly associated with clustering after Bonferroni correction, though these findings remain exploratory due to limited sample size and lack of longitudinal data.
Conclusions: Bacterial GWAS of both M. a. subsp. abscessus and M. a. subsp. massiliense revealed no direct genomic evidence of transmission. Although the Manhattan plots showed statistically significant SNPs, these findings highlight the importance of cautious interpretation of association signals and underscore the need for integrated genomic and thorough epidemiological data to elucidate transmission in the M. a. subsp. abscessus complex.