Construction And Analysis Of High-quality Pan-genome Landscape Of Bacillus Subtilis

Bacillus subtilis as widely used microbial species,is of great significance in studying microbial community relationships,adaptive evolution in various niches,engineering cell factories that produce specific products,and designing genome reduction.The pan-genome analysis is an effective method for studying the characteristics and functions of genes within species.The reliable pan-genome results usually contribute to many researches,such as rational design of bacterial genomes.However,it is still unclear how to obtain high-quality pan-genome results within certain species.In this dissertation,the old and new data sets containing different complete-or chromosome-assembly-level strains of B.subtilis were collected during one-year period.In the whole analysis process,the same analyses,programs and parameters were performed on both data sets.Then,the results from both data sets were compared separately,which showed that the ultimate B.subtilis pan-genome results were very similar for the two data sets that were obtained by removing confounding strains.And it showed that our stepwise manner of removing these confounding strains(incorrectly classified,phylogenetically distinct,engineered genome-reduced,artificial strains,strains with more pseudogenes and multiple nearly clonal laboratory strains)could significantly improve the reliability and quality of the B.subtilis pan-genome landscape.Furthermore,redundant genomes could impact judgment regarding the true pangenome state of species(open versus close).Collectively,the screening method also quantitatively and qualitatively eliminated the bias phenomenon in the pan-genome analysis to a certain extent.In total,the results of this academic dissertation will provide useful information for understanding genome function and evolution,providing a structure for reducing genome complexity through deletion of candidate genes.Most significantly,it will reduce errors in pan-genome analyses at the beginning of data processing and provide perspective on the selection of suitable strains aimed at producing representative high-quality pan-genome landscapes within species.The results could give insight into effectively reducing the time required to prepare data for pan-genome analysis,improving the efficiency of the pan-genome analysis process and contribute to the future pan-genome analysis pipeline or protocol as an important quality control step.