Truth in Science

The phrase “junk DNA” was first coined by Susumu Ohno in his Brookhaven Symposium paper, "So Much 'Junk' DNA in our Genome”. During the 1980s and 1990s, the term was increasingly used to describe all non-coding DNA sequences (at least 97% of the genome). With the completion of the human genome project, the subsequent publication in Nature, and the identification of some 22,000 protein coding genes, little insight was provided into the biological processes involved in the regulation of gene expression itself.

Over the last few years, however, the US National Institutes of Health's National Human Genome Research Institute (NHGRI) has organized the ENCODE (Encyclopedia of DNA Elements) project to identify the functional components within the human genome. Because of the immense complexity of this undertaking, ENCODE started with just 44 selected regions of DNA (about 30 million base pairs) which comprise approximately 1% of the human genome. The results of this pilot project have been published recently in Nature and in 28 companion papers published in the June issue of Genome Research.

Perhaps the most surprising finding is the fact that the vast majority of DNA (coding and non-coding) is transcribed into RNA and that these transcripts overlap each other and are taken from both strands of the double helix (sense and anti-sense). Only a very small proportion of this RNA is messenger RNA (mRNA) and actually translated into protein. The vast majority of the transcribed RNA is involved in the control of the expression of the genes themselves. In addition, the DNA from which regulatory RNA is transcribed may be very distant from the genes they control, even situated on different chromosomes.

How the abundant and varied RNA transcripts move to their targets without interfering with one another is also perplexing scientists. Some of these regulatory transcripts can be many times larger than the genes they control.

A second surprising finding involved the comparison of multispecies alignments of ENCODE regions using sequences from 22 other mammals published in Genome Research. Up to 50% of the features lacked “sequence constraint” which basically means that the sequences are very different and show little conservation across species. A picture is gradually emerging of levels of encrypted genomic information far more complex than that originally considered. Maybe the term “junk DNA” is now nothing more than junk itself. We will keep you posted.