Table 1 Challenges for metatranscriptomics, metaproteomics, and metabolomics in microbiome studies

Metatranscriptomics, metaproteomics, and metabolomics each have their own shortcomings. Metatranscriptomic experiments rely on obtaining sufficient high-quality RNA from the sample source; something which can be quite challenging due to the ubiquitous presence of RNases in host-derived samples. In addition, metatranscriptomic sequencing can often become saturated with reads from less-informative, but highly expressed transcripts (i.e., ribosomal proteins, translation factors, major outer membrane proteins) from the most abundant microbes present, obscuring the detection of functionally important, but less abundant transcripts/proteins. Therefore, the quality of RNA as well as the depth of measurement is important aspects that need to be evaluated or considered in metatranscriptomics.

Compared to metagenomics and metatranscriptomics, metaproteomics has a lower depth of measurement and can only capture 10–20% of expressed proteins in human gut microbiomes [27, 33, 34]. MS spectra can also be saturated with the highly abundant proteins from dominant species, and this issue is unlikely to be resolved by increasing the speed or time of MS scanning. However, applying off-line protein/peptide separation (such as using sodium dodecyl sulfate polyacrylamide gel electrophoresis) or targeted enrichment strategies (such as using activity-based probes [35]) may to some extent address this limitation. In addition, as metaproteomics is still in its infancy for the study of microbiomes, there is still a lack of universal guidelines and protocols for properly performing metaproteomic experiments and interpreting metaproteomic results. Therefore, careful considerations should be made for sample preparation, MS measurement, bioinformatic workflows, and data reporting (readers are directed to this perspective article for more details [36]).

The major challenge for metabolomics in microbiome studies is the difficulty to distinguish host- and microbiome-origin metabolites and directly link metabolites to specific taxa [37]. One feasible approach to address this issue is to identify co-variations between metabolites and microbial species, which is indicative for species-specific metabolite production, through integrative analysis of microbiota compositions with metabolite profiles [38,39,40,41]. Other approaches, such as protein stable-isotope probing (protein-SIP) [42], can also link the metabolism of a specific substrate to phylogenetic information by monitoring the isotopes in microbial protein sequences with mass spectrometers and may eventually aid in microbiome metabolic reconstructions.