BE: The hidden layer of noncoding RNA in the evolution and development of complex organisms.

John S. Mattick
Institute for Molecular Bioscience
University of Queensland, Brisbane QLD 4072, Australia

It appears that the genetic programming of higher organisms has been  fundamentally misunderstood for the past fifty years. Although only  1.2% encodes proteins, most of the mammalian genome is transcribed,  mainly (~98%)as non-protein-coding RNAs. These transcripts are developmentally regulated and include complex interlaced clusters of  overlapping sense and antisense transcripts, and intergenic  transcripts, that appear to participate in both local and long- distance regulatory networks. Some of these transcripts are processed  to small RNAs, including snoRNAs that direct modification of other  RNAs, and miRNAs that control many aspects of development. Most  snoRNAs and miRNAs remain to be identified, and it appears that there  may be hundreds of thousands of these and other classes of yet-to-be- discovered regulatory RNAs in mammals. In addition, a significant  proportion of the genome appears to be under selection, both positive  and negative, including thousands of ultra-conserved sequences and  transposon-free regions that are essentially unchanged in mammals.  These observations, and the many genetic phenomena shown to be RNA- directed, suggest that much, if not most, of the genomes of mammals  and other complex organisms encodes regulatory RNAs that direct the  trajectories of differentiation and development via control of  epigenetic memory, promoter selection, splicing, RNA editing, mRNA stability and translation. This conclusion is supported by  theoretical considerations and empirical data which show that  regulatory overhead rises non-linearly with complexity, and that  introns contain sequences that are conserved elsewhere in the genome  in functionally congruent networks. This system appears to  progressively dominate genomic programming as developmental  complexity increases.