The evolutionary characterisation of bacterial diversity from DNA sequence data

DNA sequence data are being increasingly used to characterise biodiversity, not least in groups of organisms in which traditional taxonomic approaches have proved of limited use. One group that is particularly dependent on DNA approaches, and particularly challenging, is the bacteria. Only a tiny fraction of bacteria are culturable and the true species richness of bacteria, as defined by current methodology, could number in the billions. However, although a wealth of sequence data for bacteria is becoming available, there remain major theoretical challenges to characterising the diversity of bacteria. First and foremost, bacteria have proved difficult to accommodate within traditional species definitions developed for plants and animals, because of important differences in their mode of inheritance. Bacteria are clonal (they reproduce by simple division of cells), yet they can exchange DNA by a variety of mechanisms, some of which occur most often between closer relatives whereas others occur between distantly related strains. However, the same basic processes cause diversification in bacteria as in plants and animals: the question is to what extent do these processes act together to produce units equivalent to species, rather than acting separately on different genes to produce a more complex pattern of diversity. On top of this problem, methods for identifying evolutionarily and biologically meaningful units of diversity from DNA data are in their infancy: most studies use crude thresholds of DNA divergence to delimit species, or graphical approaches to delimit species by eye, rather than statistical models to test for the action of different processes known to be important for causing diversity to evolve. This project will develop new methods for characterising the diversity of bacteria and use them to test whether bacteria do fall into simple units of diversity equivalent to species, or whether a more complex model of diversity is needed. First, we will develop a broadly applicable suite of new methods for identifying units of diversity from DNA sequence data. The methods will range from those suitable when only a single gene region has been sequenced from each individual, to those suitable when several genes have been sequenced from each individual, called multi-locus sequence analysis (MLSA). Software will be made freely available to enable other researchers to apply our methods in a broad range of applications. The software will be tested in relation to two existing databases, one compiling sequences of a single gene region (16S rRNA) from several hundred thousand isolates of bacteria, and one sampling bacterial genomes from environmental samples. To answer our central question concerning the simplicity or complexity of bacterial diversity, we will generate a new dataset compiling gene sequence data for the Bacillus cereus species complex. This group includes strains with beneficial roles in soil and the plant surfaces, such as nutrient cycling and blocking of plant pathogens, whereas genetically similar strains are disease agents in humans, other mammals or insects. Going beyond previous studies, we will sequence genes with important ecological functions, such as those involved in attacking host defences, as well as the so-called 'house-keeping' genes involved in basic biological processes that are normally used in MLSA studies. This will allow a more comprehensive test of different scenarios for diversification, in particular comparing functional units with different ecological attributes. The results will indicate whether simple units of diversity exist or whether the pattern of diversification is different depending on which set of genes or which aspect of diversity is being considered. The outputs will establish new methods and evidence both for practical delimitation of bacterial diversity and for theoretical debates on the evolution of diversity and the nature of species.

DNA sequence data are increasingly used to characterise biodiversity, not least in groups of organisms in which traditional taxonomic approaches have limited use. However, methods for identifying biologically meaningful units of diversity from DNA data are in their infancy. This project will develop new methods for characterising diversity in a group that is particular dependent on DNA data, and particularly challenging, namely the bacteria. We will test whether bacteria fall into simple units of diversity equivalent to species, or whether a more complex model of diversity is needed. Recent work has explored the nature of bacterial species using multi-locus sequence analysis (MLSA) of house-keeping genes. We will go beyond this by comparing patterns of diversity between core genes and those with key ecological functions, including genes on plasmids. What are the arenas of drift, natural selection and recombination for these different genes? Statistical methods testing directly for the signature of these processes will be developed from methods recently devised by the PI and collaborator (Track Record). The methods will be implemented in open source software and their use on large-scale datasets (e.g. 16S rRNA surveys) explored. Finally, we will apply the methods to a case study sampling multi-locus core and ecological genes in the Bacillus cereus species complex, a group that includes strains with beneficial roles in soil and the phytosphere and genetically similar strains that are pathogenic to humans, other mammals or insects. The results will establish new methods and evidence both for practical delimitation of bacterial diversity and for theoretical debates on the evolution of diversity and the nature of species.