In correspondence and comments on some of my blogs, there have been confusions or questions as to what I mean by "natural genetic engineering" (NGE). I will use this blog to spell out what my understanding of NGE is. Then I will discuss some implications of our knowledge of NGE for thinking about how it operates in evolution, with special emphasis on where experimental and conceptual gaps need to be filled.
For me, NGE is shorthand to summarize all the biochemical mechanisms cells have to cut, splice, copy, polymerize and otherwise manipulate the structure of internal DNA molecules, transport DNA from one cell to another, or acquire DNA from the environment. Totally novel sequences can result from de novo untemplated polymerization or reverse transcription of processed RNA molecules.
NGE describes a toolbox of cell processes capable of generating a virtually endless set of DNA sequence structures in a way that can be compared to erector sets, LEGOs, carpentry, architecture or computer programming.
NGE operations are not random. Each biochemical process has a set of predictable outcomes and may produce characteristic DNA sequence structures. The cases with precisely determined outcomes are rare and utilized for recurring operations, such as generating proper DNA copies for distribution to daughter cells.
It is essential to keep in mind that "non-random" does not mean "strictly deterministic." We clearly see this distinction in the highly targeted NGE processes that generate virtually endless antibody diversity.
In summary, NGE encompasses a set of empirically demonstrated cell functions for generating novel DNA structures. These functions operate repeatedly during normal organism life cycles and also in generating evolutionary novelties, as abundantly documented in the genome sequence record.
Some NGE functions help us to understand the mechanistic details of rapid evolutionary changes. Rapid changes include the evolution of novel proteins by reorganization of existing functional domains to generate new combinations of biochemical activities and the distribution of regulatory signals to multiple sites in the genome.
Like all biochemistry, NGE functions are subject to cell regulation (turning on and off in response to sensory inputs), and NGE operations are targetable to particular locations in the genome by well-defined molecular interactions. A surprisingly large number of external stresses and life history events activate NGE genome change operators. Many of these, such as interspecific mating, are just the kinds of unusual biological interactions that we would expect after major ecological disruption.
NGE is not an explanatory principle. In this, it differs from the use many evolutionary biologists have made of the descriptive phrase, Natural Selection, to cover gaps in their accounts of adaptive novelties. NGE is only a set of well-documented DNA change operators.
While NGE can help in understanding the molecular details of rapid and widespread genome change, it does not tell us what makes genomic novelties come out to be useful. How natural genetic engineering leads to major new inventions of adaptive use remains a central problem in evolution science. To address this problem experimentally, we need to do more ambitious laboratory evolution research looking for complex coordinated changes in the genome.
If we are able to observe cells coordinating NGE functions to make useful complex inventions in real time, major questions arise. How do they perceive what may be useful? We need to find out whether there are feedbacks between sensory inputs and genome changes. Is there any connection between the biological challenge and the NGE output? Cells can adjust other activities to meet the goals of survival, growth and reproduction. Can they do the same with DNA changes? We need to figure out how to do experiments on this.
Right now, for example, we do not even know if E. coli produces the same DNA changes in response to carbohydrate starvation (producing the internal signal cAMP) as it does in response to amino acid starvation (producing the internal signal ppGpp). There is certainly no scientific problem in postulating that high levels of cAMP may stimulate different DNA changes than high levels of ppGpp.
If experiments show that cells can make distinct appropriate NGE responses to different adaptive challenges occurs, we need to figure out how they do so. This almost certainly would prove to be more than a strictly mechanical process. How do cells carry out their computations to make useful goal-oriented responses? A successful answer to that question will certainly involve cybernetics. If such investigations take evolution science into areas that are more than strictly material, so be it. As long as we stay within the realm of natural processes, there are no boundaries on what science can address.