Where Do New Genes Come From?

Genes are the carriers of inheritable information. Yeast has just over 5,000 genes while humans have over 20,000 genes. The last common ancestor between humans and yeast lived about a billion years ago. The number of genes in this last common ancestor was most likely closer to the number found in yeast today than in humans. Thus over time the human genome, and to a lesser extent the yeast genome, has accumulated new genes. But where do these genes come from? The classical mechanism for the appearance of new genes is gene duplication. In this model a gene gets copied and over time the two copies of the gene accumulate different mutations until they no longer have the same function. But in more recent years a number of genes have been discovered that point to the existence of an alternative mechanism. 

It is now suggested that genes could pop into existence through random changes in the so called junk DNA. Junk DNA is the large part of the human genome that does not consist of genes or other DNA with known function. Such genes are called “de novo genes”.  Proteins generally are made of a combination of individual “bricks”, named protein domains, each with their own function. In the classical gene duplication model the two copies of the gene are first identical but gradually accumulate mutations resulting in slight changes to one or multiple of the domains resulting in a slightly different function. Because of this similarity between genes that originate from the same ancestral gene we can classify these genes in families. Let us compare genes to cutlery. Let us compare the ancestral protein to a knife. It is made of two functional domains, a handle and the cutting blade. When the gene gets duplicated and starts to accumulate mutations over time it can give rise to a fork. The handle domain was preserved but the cutting blade domain gradually became wider, less sharp and evolved 3 extensions. The difficulty with the de novo gene mechanism is how it is possible that a purely random sequence could lead to a functional gene.    


Read more at the Scientific American