More help arrives to predict matches for miRNAs. Credit: Katie Ris-Vicari

The small but biologically extremely influential microRNAs were again under intense scrutiny in 2008. A microRNA exerts its control by binding to the 3′ UTR of a target mRNA and prevents its translation, either by promoting mRNA degradation or by hindering the translation process.

To understand their true biological significance, the large pool of microRNAs must be matched to the even larger pool of mRNAs in a cell. Computational target predictions are notoriously error prone, and last year we called for methods to do large-scale wet-lab validation of microRNA targets predicted in silico. In this regard, 2008 brought several successful efforts. The groups of Nikolaus Rajewsky and David Bartel independently decided to wed proteomics and molecular biology techniques and used quantitative mass spectrometry to assess fluctuations in protein levels at proteome scale in cells with altered levels of a specific microRNA (Nature 455, 58–63; 2008, Nature 455, 64–71; 2008). Because the end result of regulation by a microRNA is always a reduction in protein, the targets should be identifiable with a quantitative, proteome-wide screen.

Another system-wide approach to tackle the microRNA-target matching question is degradome sequencing, introduced by the groups of Pamela Green (Nat. Biotechnol. 26, 941–946; 2008) and Michael Axtell (Curr. Biol. 18, 758–762; 2008) in plants. Here the researchers sequenced products of microRNA-mediated mRNA decay and used the resulting sequence signatures to identify the microRNA. This strategy was very successful in plants, where the match between microRNA and target is perfect; it remains to be seen how easily it can be adapted to organisms with less perfect matches.

In the meantime, the computational experts incorporated the increasing amount of experimental data into new algorithms for target prediction. One example is mirWIP, a program from the group of Victor Ambros that incorporates information from a large data set of microRNA-associated mRNAs—identified by immunoprecipitation of the RNA-induced silencing complex—to predict miRNA targets in Caenorhabditis elegans (Nat. Methods 5, 813–819; 2008).

All these system-wide efforts will improve the quality of target prediction and will provide good candidates for experimental validation—a high-throughput wet-lab technique is still needed. Stay tuned.