For those waiting on college acceptance letters these days, SAT stands for scholastic aptitude test, and the big news about them is that they recently were revamped.

For those in genomics, SAT stands for sense-antisense transcripts, and the news about them is that they are among the forms of RNA that are revamping the view of the relationship between transcription, translation and ultimate functional importance of DNA and RNA that do not code for proteins.

The original view of the relationship between DNA, RNA and proteins was that one gene codes for one protein, via one kind of messenger RNA, and that translated DNA sequences are those with the greatest biological importance, while much of noncoding DNA is "junk" with no functional relevance.

To be sure, there always has been an understanding, at least among thoughtful researchers, that there is a difference between not knowing what something's function is and asserting that it has no function. And noncoding RNA of various stripes, such as siRNA and miRNA, already has attracted a good deal of research interest in recent years.

But given the size of the human genome and technical limitations on its study, most disease research to date has nevertheless focused on the 2 percent of the human genome known to code for proteins. Furthermore, a lot of the research on noncoding RNAs has tended to be via bioinformatics methods.

SATs are one type of noncoding RNA. While it was once thought that only one strand of DNA was transcribed at any given point of the genome, SATs are pairs of RNA molecules generated from opposite DNA strands at the same spot. Essentially, each RNA is a palindrome of the other, though they are processed differently and, thus, the sequence inversion is not perfect in the final RNAs. SAT pairs have been implicated in various stages of gene regulation, including transcription, mRNA processing, splicing, stability, transport and translation.

In the April 1, 2005, issue of Genome Research, in a study titled "Disclosing hidden transcripts: mouse natural sense-antisense transcripts tend to be poly(A) negative and nuclear-localized," and jointly published by scientists from the Riken Institutes in Tsukuba, Yokohama and Saitama, Japan, and the Yokohama City University, scientists investigated such in silico SATs in a microarray study.

"The SATs identified in silico were actually transcribed, and they were transcribed differentially, depending on the cell types and tissues. Many of the SATs appear to be noncoding RNA, but still they are actually transcribed," Hidenori Kiyosawa, researcher at the Riken BioResource Center and lead author of the study, told BioWorld Today. He added that "this is the first SAT expression analysis at the genome level."

One gene type the researchers studied was imprinted genes. Whereas normally both the maternal and paternal alleles of a gene would be expressed in their offspring, in imprinted genes, one allele is silenced.

"The mechanism underlying this phenomenon is not well understood, but the imprinted genes often exist as a gene cluster on the chromosomes so they are thought to be under chromatin level regulation," Kiyosawa said.

In their study, the researchers found a number of SAT pairs within such imprinted chromosomal regions. Whether their expression is a cause or a consequence of imprinting is unclear, but the scientists suggested that antisense RNA may determine expression of the sense strand by altering its methylation status.

Another finding was that the SAT's they studied tended to not have a poly-A tail, the string of adenine nucleotides added to RNA molecules during post-translational processing to increase their stability.

Another paper, which appears in the March 24, 2005, issue of ScienceExpress, also investigated polyadenylation. That study was published by researchers from Affymetrix Inc. and the National Institutes of Health, under the title "Transcriptional Maps of 10 Human Chromosomes at 5-Nucleotide Resolution."

Polyadenylated RNA has received the most experimental attention for several reasons. For one thing, it tends to be transcribed, and, from a purely practical standpoint, its tail makes it easy to identify. However, the ScienceExpress paper, which investigated about 30 percent of the human genome using tiling microarrays, found that about 15 percent of the DNA they investigated was transcribed into RNA, and almost all of that RNA was not polyadenylated.

They also found that about a third of transcribed RNA was bimorphic - that is, it could be found in either an adenylated or a nonadenylated state. Many bimorphic RNAs were from genome regions known to code for proteins. The scientists suggest that polyadenylation itself may be one regulatory method controlling whether or not they are translated into proteins.