Transcription factors (TFs) are a well-studied set of regulatory proteins that modulate gene expression by binding to sequence-specific motifs in DNA (see The Transcription Factor Encyclopedia for a compendium of such motifs).
Analogous to genome regulation by TFs binding to DNA is the regulation of the transcriptome by RNA-binding proteins (RBPs) binding to RNA. However, at a chemical level, it could be said that RNA presents a more exciting molecule than DNA. Whereas DNA is wall-to-wall double helix, RNA bends and contorts itself into all manner of bulbous shapes, making use of its ability to form intramolecular bonds.
This difference between DNA and RNA gives RBPs more options than TFs when it comes to binding specificities. In some cases RBPs do still concern themselves with primary sequence, but others are more interested in the shape of the target RNA, exploiting the variability in RNA secondary structure.
Hisanori Kiryu and colleagues from the University of Tokyo, Japan, develop an algorithm, CapR, that rapidly calculates secondary structure probabilities base-by-base for a given set of trancriptome data. When applied to CLIP data, which is a specific kind of transcriptome data containing only those sections of RNA bound by an RBP interest, the shapes that each RBP likes to bind to become apparent.
Kiryu and colleagues explore a number of RBPs with their method. For example, they find that Pumilio-2, which regulates membrane excitability in neurons, has a preference for hairpin loop structures, while FMR1, which is associated with fragile X syndrome, preferentially binds to the internal and bulge loops of RNA.
As more data becomes available, CapR will be poised to decipher even more of the shapes that form the RBPome.
CapR: revealing structural specificities of RNA-binding protein target recognition using CLIP-seq data
Genome Biology 2014, 15:R16
Go to article >>