Since the advent of sequencing technologies,the determination of microbial diversity to predict microbial functions,which are the major determinants of soil functions,has become a major topic of interest,as evidenced ...Since the advent of sequencing technologies,the determination of microbial diversity to predict microbial functions,which are the major determinants of soil functions,has become a major topic of interest,as evidenced by the 900 publications dealing with soil metagenome published up to 2017.However,the detection of a gene in soil does not mean that the relative function is expressed,and the presence of a particular taxon does not mean that the relative functions determined in pure culture also occur in the studied soil.Another critical step is to link microbial community composition or function to the product analyzed to determine flux rates.Indeed,flux rates might not only be highly dynamic,but several metabolites can depend on different reactions,which makes the link to one process of interest difficult or even impossible.This review also discusses biases caused by sampling,storage of samples,DNA extraction and purification,sequencing(amplicon-vs.metagenome sequencing),and bioinformatic data analysis.Insights and the limits of predicting microbial interactions by network inference methods are critically discussed,and finally,future directions for a better understanding of soil functions by using measurements of microbial diversity are presented.展开更多
文摘Since the advent of sequencing technologies,the determination of microbial diversity to predict microbial functions,which are the major determinants of soil functions,has become a major topic of interest,as evidenced by the 900 publications dealing with soil metagenome published up to 2017.However,the detection of a gene in soil does not mean that the relative function is expressed,and the presence of a particular taxon does not mean that the relative functions determined in pure culture also occur in the studied soil.Another critical step is to link microbial community composition or function to the product analyzed to determine flux rates.Indeed,flux rates might not only be highly dynamic,but several metabolites can depend on different reactions,which makes the link to one process of interest difficult or even impossible.This review also discusses biases caused by sampling,storage of samples,DNA extraction and purification,sequencing(amplicon-vs.metagenome sequencing),and bioinformatic data analysis.Insights and the limits of predicting microbial interactions by network inference methods are critically discussed,and finally,future directions for a better understanding of soil functions by using measurements of microbial diversity are presented.
