MetaRibo-Seq measures translation in microbiomes

No method exists to measure large-scale translation of genes in uncultured organisms in microbiomes. To overcome this limitation, we develop MetaRibo-Seq, a method for simultaneous ribosome profiling of tens to hundreds of organisms in microbiome samples. MetaRibo-Seq was benchmarked against gold-standard Ribo-Seq in a mock microbial community and applied to five different human fecal samples. Unlike RNA-Seq, Ribo-Seq signal of a predicted gene suggests it encodes a translated protein. We demonstrate two applications of this technique: First, MetaRibo-Seq identifies small genes, whose identification until now has been challenging. For example, MetaRibo-Seq identifies 2,091 translated, previously unannotated small protein families from five fecal samples, more than doubling the number of small proteins predicted to exist in this niche. Second, the combined application of RNA-Seq and MetaRibo-Seq identifies differences in the translation of transcripts. In summary, MetaRibo-Seq enables comprehensive translational profiling in microbiomes and identifies previously unannotated small proteins.


Pearson Correlation
MetaRibo-Seq correlates to protein levels in an aerobic mock community.
Replicates reads were combined and treated as a single sample for this visualization.
(c) Histogram showing the fold enrichment of signal across coding regions flanked by UTRs (n = 2263) compared to signal across known UTR regions for RNA-Seq, MetaRibo-Seq, and Ribo-Seq in E. coli.
(a) Proportion of count data from transcriptomics, MetaRibo-Seq, and proteomics corresponding to E. coli (red) and B. thetaiotaomicron (blue) in a two member mock mixture.
(a) For Sample A, scatterplot of MetaRibo-Seq RPKM and metaproteomics NSAF both log-scaled (n=497 proteins). Replicates reads were combined and treated as a single sample for this MetaRibo-Seq demonstrates some weak signs of overall codon-resolution.
(a-d) Triplet periodicity across footprint lengths for Sample A, B, C, and D, respectively. Colors indicate which frame a read falls within. MetaRibo-Seq demonstrates stronger codon resolution in taxa-specific analyses.

Bacteroides
(a) All contigs assigned to the genera Bacteroides are considered from Sample A and B, respectively. Only these contigs are considered in triplet periodicity analyses.
(b and c) The same triplet periodicity analysis for Faecalibacterium and Alistipes, respectively. 0 50 100 translation oxidation-reduction process carbohydrate metabolic process translational elongation electron transport chain transmembrane transport metabolic process protein folding glycolytic process Number of clusters with ≥ 5 sequences