Metabarcoding is an increasingly popular and accessible method for assessing bacterial communities across a wide range of environments, and as the sequence data archives grow, sequence data reuse will likely become an important source of novel insights into the ecology of microbes. While literature on the benefits of longer read lengths for the study of microbial communities, little is known about the (re)usability of shorter (<200 bp) read lengths, but this information is essential to improve the reuse and comparability of metabarcoding data across studies. This study reanalyzed three 16S rRNA datasets targeting aquatic, animal-associated, and soil microbiomes, and evaluated how processing the sequence data across a range of read lengths affected the resulting taxonomic assignments, biodiversity metrics, and differential (i.e., before-after treatment) analyses. Short read lengths successfully recovered ecological patterns, and limited increases in resolution were observed beyond 100 bp reads across environments. Furthermore, abundance-weighted diversity metrics (e.g., Inverse Simpson index or Bray-Curtis dissimilarities) were more robust to variation in read lengths. Importantly, the total number of ASVs detected increased with read length, highlighting the need to consider metabarcoding-derived diversity estimates within the context of the bioinformatics parameters selected. This study provides evidence-based guidelines for the processing of short reads.

Disturbances alter the diversity and composition of microbial communities, but whether microbiomes from different environments exhibit similar degrees of resistance or rates of recovery has not been evaluated. Here, we synthesized 86 time series of disturbed mammalian, aquatic, and soil microbiomes to examine how the recovery of microbial richness and community composition differed after disturbance. We found no general patterns in compositional variance (i.e., dispersion) in any microbiomes over time. Only mammalian microbiomes consistently exhibited decreases in richness following disturbance. Importantly, they tended to recover this richness, but not their composition, over time. In contrast, aquatic microbiomes tended to diverge from their pre-disturbance composition following disturbance. By synthesizing microbiome responses across environments, our study aids in the reconciliation of disparate microbial community assembly frameworks, and highlights the role of the environment in microbial community reassembly following disturbance.