PSB 2010 Tutorial: Computational Studies of Non-coding RNAs

The central dogma of molecular biology characterizes RNA as a simple working copy of DNA, simply transporting a code from the genome into the protein biosynthesis machinery [14, 15]. However, recent discovery of RNA interference (RNAi) [25, 26], the post transcriptional silencing of gene expression via interactions between mRNAs and their regulatory RNAs, has drastically changed the picture that is portrayed by the central dogma. In the new picture, non-coding RNAs (ncRNAs) play a significant regulatory role in the cell. FANTOM and ENCODE genome annotation studies have revealed that a large fraction of the genome sequences give rise to ncRNAs [24,58]. A recent computational screen estimated the number of small regulatory RNAs, which form an important class of non-coding RNAs, in Arabidopsis thaliana to be in the order of 75,000 [61]. Among small RNAs, two subclasses form the bulk of all regulatory RNAs: microRNAs (miRNAs) and small interfering RNAs (siRNAs) — which are of similar length (21 to 25 nt) and composition but different by origin. It is predicted that these two subclasses regulate at least one-third of all human genes. There are many other classes of non-coding RNAs with functionalities beyond simple regulation of gene expression: examples include snoRNAs, snRNAs, gRNAs, and stRNAs, which respectively perform ribosomal RNA (rRNA) modification, RNA editing, mRNA splicing and developmental regulation [32]. Even for these well-studied RNAs, their precise mode of function remains poorly understood. In addition to such endogenous ncRNAs, antisense oligonucleotides have been synthesized as exogenous inhibitors of gene expression. Antisense gene silencing technology is currently used as a research tool and for therapeutic purposes. The therapeutic objective of antisense technology is to block the production of disease-causing proteins. In principle, these artificial regulatory RNA molecules could be employed as drugs for the treatment of a variety of human diseases including various types of cancer, rheumatoid arthritis, brain diseases, and viral infections [27]. As a research tool, antisense nucleic acids may be used to study metabolic networks by controlling or interfering with the dynamics and function of various modules in the network. Furthermore, synthetic nucleic acid systems have been engineered to self-assemble into complex structures performing various dynamic mechanical motions [33, 62, 66]. Despite advances in computational studies of non-coding RNA, there are still many open areas and unresolved issues particularly for high-throughput applications based on the new genome sequencing technologies.