A New Genome-to-Genome Comparison Approach for Large-Scale Revisiting of Current Microbial Taxonomy

Microorganisms

Yen-Yi Liu¹  Von-Wun Soo²  Chih-Chieh Chen³  Ming-Hsin Tsai⁴ 

[1] Central Regional Laboratory, Center for Diagnostics and Vaccine Development, Centers for Disease Control, Taichung 40855, Taiwan;Department of Computer Science, National Tsin Hua University, Hsinchu 30013, Taiwan;Institute of Medical Science and Technology, National Sun Yat-sen University, Kaohsiung, 80424, Taiwan;Institute of Population Health Sciences, National Health Research Institutes, Miaoli County 35053, Taiwan;

关键词: microbial taxonomy;    whole genome comparison;    bacterial classification;    bacterial identification;   

DOI  :  10.3390/microorganisms7060161

来源: DOAJ

【 摘 要 】

Microbial diversity has always presented taxonomic challenges. With the popularity of next-generation sequencing technology, more unculturable bacteria have been sequenced, facilitating the discovery of additional new species and complicated current microbial classification. The major challenge is to assign appropriate taxonomic names. Hence, assessing the consistency between taxonomy and genomic relatedness is critical. We proposed and applied a genome comparison approach to a large-scale survey to investigate the distribution of genomic differences among microorganisms. The approach applies a genome-wide criterion, homologous coverage ratio (HCR), for describing the homology between species. The survey included 7861 microbial genomes that excluded plasmids, and 1220 pairs of genera exhibited ambiguous classification. In this study, we also compared the performance of HCR and average nucleotide identity (ANI). The results indicated that HCR and ANI analyses yield comparable results, but a few examples suggested that HCR has a superior clustering effect. In addition, we used the Genome Taxonomy Database (GTDB), the gold standard for taxonomy, to validate our analysis. The GTDB offers 120 ubiquitous single-copy proteins as marker genes for species classification. We determined that the analysis of the GTDB still results in classification boundary blur between some genera and that the marker gene-based approach has limitations. Although the choice of marker genes has been quite rigorous, the bias of marker gene selection remains unavoidable. Therefore, methods based on genomic alignment should be considered for use for species classification in order to avoid the bias of marker gene selection. On the basis of our observations of microbial diversity, microbial classification should be re-examined using genome-wide comparisons.

【 授权许可】

Unknown