With the increasing adoption of discovery -omics in the life sciences, a large number of analysis tools for differential expression analysis (DEA) have been introduced over the years. While such tools tend to be developed with one particular -omics modality in mind, they can often be applied across technologies to solve common issues. This is particularly the case when -omics data share statistical and distributional properties. Herein, we showcase how tools originally developed for transcriptomics analysis are especially well-suited to solving problems in discovery proteomics and metabolomics. Using data from our own experimental work as examples of real-world implementation, we demonstrate how these methods can be used to tackle common DEA issues, such as variable sample quality, hidden batch effects, normalization, and small sample size. We believe this can be useful to novices and seasoned practitioners alike by expanding their toolkits. As multiomic and integrative analyses become commonplace, it is especially useful to capitalize on the similarities of otherwise different -omics.
Dohm-Hansen et al. (Fri,) studied this question.