Are short‐read amplicons suitable for the prediction of microbiome functional potential? A critical perspective. Issue 3 (4th July 2022)

• Record Type: Journal Article
• Authors: Heidrich, Vitor; Beule, Lukas
• Is Part Of: IMeta. Volume 1:Issue 3(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-07-04
• Subjects: compositional data -- functional potential profiles -- intragenomic marker gene variability -- microbial functions -- microbiome data -- short‐read amplicon sequencing -- taxonomic resolution; Metagenomics -- Periodicals; Bioinformatics -- Periodicals; Bioinformatics; Metagenomics; Metagenomics; Metagenome; Computational Biology; Periodicals; Periodical; 576.5
• DOI: 10.1002/imt2.38
• Languages: English
• ISSNs: 2770-596X

Abstract

Abstract: Taxonomic marker gene analysis allows uncovering taxonomic profiles of microbial communities at low cost, making it omnipresent in microbiome research. There is an ever‐expanding set of tools to extract further biological information from this kind of data. In this perspective, we enunciate several concerns regarding the biological validity of predicting functional potential from taxonomic profiles, especially when they are generated by short‐read sequencing. The taxonomic resolution of marker genes, intragenomic variability of marker genes, and the compositional nature of microbiome data are discussed. Combining actual measurements of microbiome functions with predicted functional potentials is proposed as a powerful approach to better understand microbiome functioning. In this context, the significance of predicted functional potentials for generating and testing hypotheses is highlighted. We argue that functions of microbiomes predicted from microbiome DNA read count data generated by short‐read amplicon sequencing should not serve as the only basis to draw biological inferences. Abstract : We raise concerns regarding the biological validity of predicting functional potential from microbiome taxonomic profiles generated by short‐read amplicon sequencing. We reason that predicted functional potential profiles can be improved by employing long‐read sequencing technologies, which in combination with independent measurements of actual functions can constitute a powerful approach to understanding microbiome functioning. Highlights: Concerns regarding the biological validity of predicting functional potential from taxonomic profiles of microbiome data sets are enunciated. Taxonomic resolution of marker genes, intragenomic variability of marker genes, and the compositional nature of microbiome data are discussed. Combining measurements of actual functions and predicted functional potential profiles is a powerful approach to understanding microbial functioning.