Computer-Based and Theoretical Identification of Regulatory RNA

Abstract

The long list of examples in this book underscores the roles that RNA (regulatory) motifs play in controlling the genetic repertoire of cells and developing organisms. As explained in Chapter 3, identification of an RNA-processing signal, a ribozyme, an RNA localization signal or a translational or mRNA stability control element (examples for each are summarized in Chap. 2, Table 2.2) allows the search for other RNA sequences that bear similar regulatory signals (consensus search). Whilst DNA regulatory elements can often be described by a consensus sequence, RNA signals are frequently composed of combined sequence and structure motifs. A sequence motif is, for example, the hexameric polyadenylation signal AAUAAA. Other RNA motifs are often characterized by complex secondary and tertiary structures that can constitute the major part of the signal, e.g. the clover-leaf pattern typical for the secondary structure of tRNAs or stemloop structures involved as a second part of the signal in delineating the polyadenylation site in several retroviridae. The search for such motifs and signals is an exercise in understanding RNA structures and sequences as a type of language. As in a human language, both single characters (nucleotides) and higher-order structures (stem-loop structures and more complex tertiary interactions such as pseudoknots) are important. This can also be summarized as interfering contexts of regulatory sequence elements (Trifonov 1996). This chapter first illustrates how searches for new RNA motifs are conducted by computer-aided database screening of genomic sequences and databases for such motifs composed of different sequences and structures. After this we will examine more general (template-free) types of searches.