Small, untranslated RNAs are present in many different organisms, ranging from bacteria to mammalian cells.  These RNAs carry out a variety of biological functions.  Many of them act as regulators of gene expression at a post-transcriptional level, either as antisense RNAs, by binding to complementary sequences of target transcripts, or by interacting with proteins.  Regulatory RNAs are involved in the control of a large variety of processes such as plasmid replication, transposition in pro- and eukaryotes, phage development, viral replication, bacterial virulence, global circuits in bacteria in response to environmental changes, or developmental control in lower eukaryotes.
At the time I began my studies there were 10 known small RNAs in Escherichia coli. Small RNAs are difficult to detect by traditional computational methods, since they have no open reading frame and the only signals you can rely on are transcriptional signals. We used a new computational approach to predict novel sRNAs and than passed our predictions to Liron Argaman and Dr. Shoshy Altuvia from the Hebrew University and to Dr. Jörg Vogel and Dr. E. Gerhart H. Wagner form the Uppsala University for lab testing.

The computational Algorithm:
Since small RNA need to be transcribed we expected them to have transcriptional signals. When investigating the 10 known small RNAs we discovered that they are all located in regions without annotated open reading frames on either of the two DNA strands. We also found that 9 of the 10 known small RNAs are conserved in other bacteria.
Our algorithm therefore is composed of the following steps:
1) extract the "empty" regions (regions without an ORF or a known small RNA, tRNA or rRNA gene, on either of the two DNA strands) - this was done based on the Colibri database.
2) Search the empty regions of the genome for transcriptional signals - We searched for sigma70 promoters and Rho-independent terminators.
3) extract sequences with a predicted terminator 50-400 bases downstream to a predicted promoter.
4) check sequences for conservation. select only sequences with good conservation

24 candidates were predicted. 23 were checked in the lab and 17 were shown to be true small RNAs, 14 of which were characterized.

A schematic representation of the Escherichia coli circular genome with locations of the small RNAs. The previously known small RNAs are colored orange. Novel small RNAs are colored red.