Statistical approaches for differential expression analysis in metatranscriptomics

Abstract

Motivation Metatranscriptomics (MTX) has become an increasingly practical way to profile the functional activity of microbial communities in situ. However, MTX remains underutilized due to experimental and computational limitations. The latter are complicated by non-independent changes in both RNA transcript levels and their underlying genomic DNA copies (as microbes simultaneously change their overall abundance in the population and regulate individual transcripts), genetic plasticity (as whole loci are frequently gained and lost in microbial lineages) and measurement compositionality and zero-inflation. Here, we present a systematic evaluation of and recommendations for differential expression (DE) analysis in MTX. Results We designed and assessed six statistical models for DE discovery in MTX that incorporate different combinations of DNA and RNA normalization and assumptions about the underlying changes of gene copies or species abundance within communities. We evaluated these models on multiple simulated and real multi-omic datasets. Models adjusting transcripts relative to their encoding gene copies as a covariate were significantly more accurate in identifying DE from MTX in both simulated and real datasets. Moreover, we show that when paired DNA measurements (metagenomic data) are not available, models normalizing MTX measurements within-species while also adjusting for total-species RNA balance sensitivity, specificity and interpretability of DE detection, as does filtering likely technical zeros. The efficiency and accuracy of these models pave the way for more effective MTX-based DE discovery in microbial communities. Availability and implementation The analysis code and synthetic datasets used in this evaluation are available online at http://huttenhower.sph.harvard.edu/mtx2021. Supplementary information Supplementary data are available at Bioinformatics online.

Controlling taxa abundance improves metatranscriptomics differential analysis

“…To address this issue, Biobakery 3 [10] takes advantage of the paired metagenomics information measured in the same individuals, and include DNA abundance as a covariate in a linear-mixed effect model to address the strong correlation between DNA and RNA abundance. A recent study [11] benchmarks the performance of six DE models controlling for either DNA or taxonomic abundances, and concludes that the model that only controls for DNA abundance has the preferred performance.…”

“…For each feature (gene pathway in a species), only samples that have all positive values in RNA, DNA, and taxonomic abundances are kept. Such filtering addresses the excessive amounts of zeroes in RNA abundances [10], and similar approach has been shown to have superior performance [11].…”

Localized coevolution between microbial predator and prey alters community-wide gene expression and ecosystem function

“…This regularization produces log2 fold changes of read counts for each sample over an intercept, and each gene is analyzed independently. To account for potential confounding variation in underlying gene copy number (i.e., due to shifts in species abundance) between samples, we included a customized normalization factor for within-taxon sum scaling and an amplicon-based estimate of the source taxon's relative abundance as a model covariate [62]. Thus, any variation in gene expression due to the variation in underlying DNA template is controlled for in our results.…”

Microbiome-Mediated Immune Signaling in Inflammatory Bowel Disease and Colorectal Cancer: Support From Meta-omics Data

“…Furthermore, due to the dynamic and compositional nature of microbiomes, many of the techniques described above require large sample sizes and result in even larger datasets that must be properly assessed for quality before their interpretation ( Zhang et al, 2021a ). Accurate reference-based taxonomic assignment of sequences depends on the quality of database that is used and achieving statistical significance, in differential expression analysis for example, can be challenging ( Zhang et al ., 2021b ). There is considerable interest in evaluating individual GI microbiomes within a clinical setting to aid screening and diagnosis using a personalized approach ( Knox et al, 2019a ).…”