Genome-wide analysis of differential translation and differential translational buffering using anota2seq

Genome-wide analysis of mRNA translation (translatomics) can improve understanding of biological processes and pathological conditions including cancer. Techniques used to identify the effects of various stimuli on the translatomes such as polysome- and ribosome-profiling necessitate adjustment for changes in total mRNA levels to capture bona fide alterations in translational efficiency. In addition to changes in translational efficiency, which affect protein abundance, translation can also act as a buffering mechanism whereby protein levels remain constant despite changes in mRNA levels (translational buffering). Herein, we present anota2seq, an algorithm for analysis of translatomes, which is applicable to DNA-microarray and RNA sequencing data. In contrast to anota2seq, current methods for analysis of translational efficiency using RNA sequencing data as input identify false positives at high rates and/or fail to distinguish changes in translation efficiency from translational buffering when two conditions (e.g. stimulated and non-stimulated cells) are compared. Moreover, we identify data- acquisition specific thresholds, which are necessary during anota2seq analysis. Finally, we employ anota2seq to show that insulin affects gene expression at multiple levels in a largely mTOR-dependent manner. Thus, the universal anota2seq algorithm allows efficient interrogation of the translatomes, including distinction between the changes in translation efficiency and buffering.