A drift‐barrier model drives the genomic landscape of a structured bacterial population

Abstract

Bacterial populations differentiate over time and space to form distinct genetic units. The mechanisms governing this diversification are presumed to result from the ecological context of living units to adapt to specific niches. Recently, a model assuming the acquisition of advantageous genes among populations rather than whole genome sweeps has emerged to explain population differentiation. However, the characteristics of these exchanged, or flexible, genes and whether their evolution is driven by adaptive or neutral processes remain controversial. By analysing the flexible genome of single-amplified genomes of co-occurring populations of the marine Prochlorococcus HLII ecotype, we highlight that genomic compartments-rather than population units-are characterized by different evolutionary trajectories. The dynamics of gene fluxes vary across genomic compartments and therefore the effectiveness of selection depends on the fluctuation of the effective population size along the genome. Taken together, these results support the drift-barrier model of bacterial evolution.

Pangenome sequence evolution within human gut microbiomes is explained by gene-specific rather than host-specific selective pressures

“…Consistent with previous estimates across multiple kingdoms of life 22 , we observe that θ π and θ w distributions across samples span 3 to 4 orders of magnitude ( Figure S4 ). Also consistent with previous estimates in bacteria over different time scales 3,7,23 , dN/dS tends to be less than one, suggesting the predominance of purifying selection at the protein level ( Figure S4 ). Our estimates of these population genetic metrics from metagenomic data are thus within an expected range and appear to behave as expected.…”

“…In this context, it is not surprising that simulations identified HGT rate, but not selective coefficients, as an important driver of molecular evolution. This model seems to fit some other bacterial genomic datasets 11,23 but awaits formal testing. Finally, we suggest that future work on pangenome evolution should try to understand what factors control shifts in the drift-selection balance and its interplay with species ecology ( N e , species lifestyle, etc.)…”

Mobile Gene Sequence Evolution within Individual Human Gut Microbiomes Is Better Explained by Gene-Specific Than Host-Specific Selective Pressures

“…S4, Supplementary Material online). Also consistent with previous estimates in bacteria over different time scales ( Sela et al 2016 ; Gardon et al 2020 ; Garud and Pollard 2020 ), dN/dS tends to be less than one, suggesting the predominance of purifying selection at the protein level (supplementary fig. S4, Supplementary Material online).…”

“…This is supported by our relatively short-term simulations, in which the HGT rate (which increases the census population size and thus N e ) was found to be a major determinant of sequence evolution, whereas gene-specific selection coefficients were not. These findings are broadly consistent with a drift-barrier model of evolution ( Bobay and Ochman 2018 ; Gardon et al 2020 ), which could be explored in additional genomic data sets and simulations. We suggest that future work on pangenome evolution should ask what factors control shifts in the drift-selection balance and its interplay with species ecology ( N e , species lifestyle, etc.)…”

How many ecological niches are defined by the superabundant marine microbeProchlorococcus?

“…The accuracy of current models (1, 21-23) might be increased by explicitly recognizing the ecological and physiological substructure of Prochlorococcus (14). Understanding the molecular evolution of Prochlorococcus depends on appropriate definitions of genetic populations, but studies have used widely differing ones (12, 19, 24, 25). Defining ecotypes in superabundant microbes presents several novel challenges.…”