When a group of researchers from Sun Yat-sen University in China published a paper last April detailing their efforts to genetically modify human embryos, responses ranged from shock that someone went ahead with a plan to change the human genetic code — even in damaged embryos that were never going to be implanted and carried to term — to fascination with the powerful technology that we've discovered that allows us to rework the building blocks of life.
But there was also another sort of reaction, and that was relief. Even though they did successfully make changes to the embryos' genomes, these changes were for the most part, wildly inaccurate: the wrong segments of genetic code were often deleted, in some areas both desired and undesired changes were made, and in some cases the changes didn't even really take effect at all.
The collective exhale was based on the reasoning that if scientists can't make these kinds of changes accurately, then we're not yet at the point where we need to worry about designer babies and a world where wealthy humans circumvent evolution and re-shape our species for their own purposes (for those who are worried about that).
But here's the thing. According to multiple researchers Tech Insider has spoken with, the Sun Yat-sen team's results could have been a lot more accurate. There are (and were) far more accurate versions of the gene-editing tool they used (CRISPR/Cas-9), and many researchers have been able to edit cells and even animal embryos with almost zero unwanted or unexpected changes.
They could have done it better, and many of those errors that people were relieved to see could have been avoided.
"It's clear they didn't know what they were doing," Dr. Scott Fahrenkrug tells Tech Insider. Fahrenkrug is the CEO of a Minnesota company called Recombinetics that has developed a way to edit the genes of animals to change specific traits, both for agricultural and biomedical uses.
"We're directly editing livestock embryos with great accuracy," says Fahrenkrug. "The technologies are so precise that they cause less off-targeting than the natural mutation rate." In other words, he's arguing they are able to know exactly what changes will occur when they make edits, and that the mutations that naturally occur cause more unwanted changes than their genome-editing process.
Harvard geneticist George Church, one of the pioneers of using CRISPR to edit DNA, told Tech Insider in an email that there had already been "improvements in the technology that greatly reduce off-target mutations that [the researchers involved in the study] cited ... but did not employ."
There were better, more accurate versions of the technology already out there.
It's not clear why the researchers opted to publish the results they did — they may have wanted to be first, and knew they'd make a splash even with their low rates of accuracy. As Nature News reported, at least four groups in China were working on similar research.
And the fact that more accurate versions of the technology exist doesn't mean they would have worked perfectly in human embryos. It's likely that that years of experimentation will be needed to come up with a version of CRISPR that works perfectly, or even very well, in humans.
But that time when people are able to accurately make genetic changes in humans may not be as far away as we think.
