Nobody’s perfect—but science can finally explain why

INTERNAL ERROR: Please contact support. You wouldn’t expect to get any errors in evolution if every possible one was eliminated over millions of years. Unfortunately, that isn’t what ended up happening with human DNA. Sorry, Darwin.

There might be 7.8 billion of us on the planet, but that is absolutely nothing compared to microbes that self-replicate so fast we’d need multiple Earths if that many humans were on the scene. Human genes are also extremely complex. Weird DNA can lurk between our genes, having many clunky and inefficient effects, as opposed to something like yeast whose genes make just one thing very efficiently. Professor Laurence Hurst and his team from the University of Bath are now finding out why we have all sorts of glitches in our DNA.

Humans and other species with relatively small population sizes have not been able to sustain a perfect genetic state through evolution. If we were single-celled organisms with the same faulty DNA, we’d have been Darwinized ages ago.

“If a mutation is a little bit bad, selection will eventually remove that mutation if given long enough,” Hurst, who co-authored a study recently published in Molecular Biology and Evolution, “But in small populations, there are fewer lucky breaks needed for the mutation to become really common and less time available to get rid of these bad mutations.”

Think of an electronics display where all the shiniest, newest computer models are all playing the same music video in super HD up front. Those would be things like algae. Human DNA would be in the used pile. Our programming is plagued with errors, because when populations are introduced to mutations that can be detrimental to them, they can often change by chance. Think of a new but really complicated computer program (remember when the now-obsolete Windows XP kept crashing?). XP had many upgraded features, but when these more complex operations are not carried out perfectly, it flashed the dreaded blue screen of death.

Programming in microbial species that have huge populations and multiply fast tends to be simpler, but much more efficient. This is because of the code written into their cells. DNA makes proteins in strings that are read by cells. The protein at the end, the stop codon, is what tells the cell to stop reading. There are three types of stop codons. Any one of these can be in any random gene, but the TAA codon is superior to the TGA and TAG codons when it comes to switching off cell programs. So wouldn’t it be more logical for natural selection to push the TAA codon as species evolve?

“Overall, species with small populations are expected to be more prone to decay and a reduced efficacy of selection,” said Hurst. “This is known as the nearly neutral hypothesis. Also, when genes make proteins we make a large number of versions of the RNA transcripts, many of which cannot be used.”

No offense, but human DNA often ends up producing junk transcripts. Under 2% of our DNA codes for protein. Whatever genetic switches are flipped to turn genes on and off also experience a high level of decay. Defective genes can be expressed for any reason, sometimes just because the next gene over is functional. The TAA codon is much more consistent in a massive population of algae. What happens by chance hardly matters in fast-multiplying microorganisms, so natural selection deleted mutations likely to glitch, which made TAA the dominant codon. So would it be possible to reprogram human DNA that way?

Hurst believes that it might be possible someday, but we aren’t technologically there yet. Gene editing has done some unbelievable things that only existed in science fiction a decade ago. The problem is that upgrading human genes to use TAA would require germline editing, which would make sure that the offspring and entire genetic line of the person being edited would also inherit the TAA advantage. This type of editing still has too many safety hazards to be approved. Parts of DNA that weren’t meant to be edited could still end up affected by the procedure, and scientists really don’t want to go there. However, Hurst’s work is still a breakthrough that will undoubtedly influence how we level up genes in the future.

“We have provided evidence that human DNA struggles to maintain the best state, thus equating what is most common and what is best shouldn’t be done,” he said. “More specifically, when designing replacement genes for gene therapy, we suggest using TAA as the stop even though it isn’t the most common in human DNA.”

Don’t get too discouraged at the thought of our genes acting like a computer that constantly malfunctions. For us to have stuck around this long, there is something evolution must have gotten right.