The impact of frequently neglected model violations on bacterial recombination rate estimation: a case study in Mycobacterium canettii and Mycobacterium tuberculosis.


Mycobacterium canettii (MCAN) is a causative agent of tuberculosis in humans, along with the members of the Mycobacterium tuberculosis complex (MTBC). Frequently used as an outgroup to the MTBC in phylogenetic analyses, MCAN is thought to offer the best proxy for the progenitor species that gave rise to the complex. Here, we leverage whole genome sequencing data and biologically relevant population genomic models to compare the evolutionary dynamics driving variation in the recombining MCAN with that in the non-recombining MTBC, and discuss differences in observed genomic diversity in the light of expected levels of Hill-Robertson interference. In doing so, we highlight the methodological challenges of estimating recombination rates through traditional population genetic approaches using sequences called from populations of microorganisms, and evaluate the likely mis-inference that arises owing to a neglect of common model violations including purifying selection, background selection, progeny skew, and population size change. In addition, we compare performance when full within-host polymorphism data is utilized, versus the more common approach of basing analyses on within-host consensus sequences.

MIDAS Network Members

This site is registered on as a development site.