We were taught wrong. The old elementary school science lesson that our DNA sequence is a 50-50 split of our mother and father omitted the salient matter that there are new, so called "de novo" mutations that spontaneously occur and are a big deal.
It wasn't until this past year that we could sequence whole human genomes of families and even single sperm cells to directly quantify how frequent these de novo mutations arise. Each of us has about 80-100 changes in our native (germline) DNA that are not found in our parent's DNA. But the source is the genomic instability from their eggs or sperm. On average about 15-20 "spelling errors" come from our mother and ~30-60 come from our father. We have a new appreciation of the father's biologic clock, with aging dads having more sperm DNA instability and increasing evidence that this phenomena is linked with a higher risk of autism and schizophrenia.
While the new mutations are rare in any given individual—representing a tiny fraction of <0.001% of one's genome—the chance of them doing harm is great. That's because they are not subject to natural, evolutionary selection. While there's a small chance the mutation could have a positive impact, the overwhelming likelihood is for a deleterious effect as we've seen with the recent studies of such mutations in kids with severe intellectual disability and other neurodevelopmental diseases.
So to respond to the 2013 Edge question, what's the worry? We should be concerned that this genomic instability in our germline DNA, and also our somatic (body cell) DNA is on the rise. We're seeing more new cases of cancer, which represents DNA off the tracks, prototypic of genomic instability. And while the aging father trend is a clear and global phenomenon, and may contribute to a small part of the increased incidence of autism, the story may be much bigger than that. So far we only know there exists a relationship with easy to diagnose traits like schizophrenia and severe cognitive disability. What about the more subtle impact of such mutations on other conditions, such as mild cognitive impairment or susceptibility to diabetes? The question one really needs to ponder is precisely why genomic instability is increasing with age and why more people are getting diagnosed with cancer from year to year—about 1.7 million Americans in 2012 and an increased incidence for 7 of the leading types of cancer, even adjusted for the advancing age of the population.
I think a significant portion of genomic instability is due to environmental effects. For example, exposure to increased radiation is a prime suspect—be it man-made thermal radiation from atmospheric greenhouse gases, or via medical imaging that uses ionized radiation. There are probably many other environmental triggers in our "exposome" that have yet to be unraveled, such as the interaction of our native DNA with our gut microbiome, or the overwhelming, pervasive exposure that we have to viruses that can potentiate genomic instability.
Although we now have an appreciation for the frequency of de novo mutations due to the spectacular advances in sequencing technology and analytics, we do not have even a rudimentary understanding of what induces them in the first place—and what are the more subtle impacts that may track with a sort of devolution of man. Especially disconcerting is that the signals of increased genomic instability are occurring in a relatively short span of time, in the context of human evolution over millions of years.
Thus, it may take a long time for this to play out. But we could do something about this by doing an in-depth study of de novo mutations and environmental interactions among hundreds of thousands of individuals and their offspring. Not that we want to suppress all de novo mutations of man, but perhaps someday there will be a way forward to prevent or screen out deleterious ones or, on the other hand, foster those which may prove favorable. It might be considered by some to represent unnatural selection, but that's what it may take to turn the tide.