It provides experimental evidence showing that most human genes have antisense transcripts.

“The availability of LongSAGE data opened the way for using SAGE tags to identify antisense transcripts.”

The observation was based on the LongSAGE data. Compared with the expression data sets detected by other approaches such as the full-length mRNA and EST, SAGE data provides a far wider and deeper mRNA detection due to its higher sensitivity and productivity.

Furthermore, the LongSAGE tag (21 bp) over the classical SAGE tag (14 bp), allows it to map to the human genome reference sequences with high specificity. By mining the LongSAGE data, mapping to human genome sequences and comparing these with existing antisense data, the study reached a new level of antisense determination.

Antisense transcripts are widely present for most human genes. Antisense transcripts can be a general means for gene expression regulation.

We have been working on SAGE-based transcriptome studies. We frequently observe that substantial SAGE tags from any given study are mapped to the known genes, but in antisense orientation. We know that SAGE is more sensitive than any other approaches for mRNA detection; therefore, SAGE data should contain more antisense information.

A particular problem in using SAGE data for genome annotation is that the classical SAGE tags have low specificity of mapping to the genomic DNA due to their short length. With the increased length of LongSAGE tags, the mapping specificity is substantially improved. The availability of LongSAGE data opened the way for using SAGE tags to identify antisense transcripts.

With the development of next-generation DNA sequencing technologies, mRNA detection will be much wider and far-reaching. Mining the new information will identify more antisense transcripts, perhaps even reaching a level where all genes will have antisense transcripts.