Deciphering Mycobacterium tuberculosis lineages through multi -omic integration
Ashleigh Cheyne1, Agnieszka Broda1, Dr Nitya Krishnan1, Dr Brian Robertson1, Dr Myrsini Kaforou2, and Dr Gerald Larrouy-Maumus1
MRC Centre for Molecular Bacteriology and Infection, Imperial College London1 Department Infectious Disease, Faculty of Medicine, Imperial College London2
The majority of tuberculosis research utilises a handful of reference Mycobacterium tuberculosis (Mtb) strains, ignoring the 7 genetically diverse phylogenetic lineages that exist worldwide. Several studies have identified differences in phenotypic features in these lineages such as virulence and antibiotic resistance, suggesting the need to study these individually and not rely on a single reference strain of Mtb. Previous literature aiming to understand these phenotypic differences has identified SNPs which are specific to certain Mtb lineages. However, the downstream phenotypic effects of these SNPs have not been fully elucidated, so it is unclear which of these SNPs are the cause of lineage-specific variations in virulence and other phenotypic factors. Here, we interrogate the functional consequences of SNPs identified in several Mtb lineages through genomics and metabolomics analysis. We have developed a novel integration analysis pipeline to analyse these datasets both individually and together through correlation analysis. This pipeline is easy to perform on metabolomics and genomic datasets and can be used on a small sample size. By applying this new pipeline, we have identified differences between lineages in several downstream metabolic pathways, potentially driven by specific SNPs which correlated with these pathways. We found that ergothioneine, which is important for redox regulation in Mtb, was significantly reduced in strains from lineage 1 compared to all other lineages in our analysis. Furthermore, we identified several SNPs which may be driving this change, including virulence factors such as WhiB6. Overall, we have uncovered new links between Mtb SNPs and metabolic pathways using a novel integration pipeline.