To the editor:

The commentary by Stephan Beck, Alexander Olek, and Jörn Walter (Nat. Biotechnol. 17, 1144, 1999) emphasizes the importance of epigenetic information in the human genome, and points out that a European alliance, the Human Epigenome Consortium (HEC), has been formed. This important initiative is to be welcomed, although it is long overdue.

It is now 25 years since it was suggested that DNA methylation is likely to have a role in the control of gene expresssion in higher organisms. Within the past 10 years, considerable evidence for this has accumulated. Yet in the whole history of the Human Genome Project, there has been almost no discussion of the distinction between cytosine and 5-methylcytosine in the genome sequence.

The authors of the Commentary refer to an organism's “epigenotype.” However, the true situation is that an organism has the same genotype in all its cells, but different types of cells and tissue have different epigenotypes. The epigenotype is the normally stable and often heritable genotype upon which additional information has been imposed. Thus, organisms have many different epigenotypes in their cells and tissues. In 1995, I wrote an article, “DNA methylation in eukaryotes: 20 years on,” which concluded as follows1:

Finally, it is important to consider DNA methylation in the context of the sequencing of the human genome. There are not four bases in human DNA, but at least five, and very likely others2. To learn the position of every cytosine, but not to know whether it is methylated or not, will clearly fail to answer many important questions about the human genome. In particular, it will not address the question of those epigenetic controls in human DNA that are due to base modification. The sequence of the genome is one level of information and can be said to comprise the full genotype of our species. The time has come to recognize the importance of the epigenotype, both in the germline cells and in the somatic tissues. When the human genome is finally sequenced there will remain the enormous challenge of deciphering the epigenetic code that is superimposed on the four-base genetic code. This will surely occupy the attention of many investigators in the next 20 years.