Five years ago I wrote for the GSA weblog, Read/write access to your genomes? Using the past to jump to the future. The basic idea is that whole-genome analysis will start to become “normal” in the 2010s, while “writing” to the genome (through CRISPR) will only start to hit the mainstream in the 2020s.
For humans, where it is going to start and become ubiquitous over time is in Mendelian diseases. Basically, those diseases characterized by a large effect mutation that genetic engineering can target. If you have a genetics education, you know that sickle cell anemia is one of those diseases. Often, it is a classic example of heterozygote advantage. Those with one copy of the mutation are more resistant to malaria, while those with two copies exhibit disease.
Last year a young woman received treatment to “fix” the mutation. NPR has been following her, and the results are good, 1st Patients To Get CRISPR Gene-Editing Treatment Continue To Thrive:
Gray and the two other sickle cell patients haven’t had any complications from their disease since getting the treatment, including any pain attacks or hospitalizations. Gray has also been able to wean off the powerful pain medications she’d needed most of her life.
Prior to the treatment, Gray experienced an average of seven such episodes every year. Similarly, the beta thalassemia patients haven’t needed the regular blood transfusions that had been required to keep them alive.
“It is a big deal because we we able to prove that we can edit human cells and we can infuse them safely into patients and it totally changed their life,” says Dr. Haydar Frangoul at the Sarah Cannon Research Institute in Nashville. Frangoul is Gray’s doctor and is helping run the study.
Wait for the baldness cure industry to enter the game, the technology will gain mainstream in a heartbeat.
I’m happy about the medical usage and don’t get why some people can’t embrace it – including for germline correaction to prevent pathological conditions and suffering.
But at the same time I’m somewhat concerned about its potential for bioweapons and personalised poisoning, attacks and assassinations. I think this can get more dangerous or at least as dangerous as nuclear weapons, but much harder to prove if being used and to keep under control.
Being able to get rid of bad, typically rare or ultra-rare Mendelian disease inducing / large phenotype effect alleles seems potentially particularly really important for “genetic rescue” of species that have been hit by falling population sizes and suffered bottlenecks (including but not exclusively your charismatic megafauna and the like). Because in that case, recovering a lot of the lost genetic diversity across the genome is maybe not so important, you really mainly just want to eliminate the bad variants that have become fixed or close to fixture in the population. If the problem with all your Pandas or whatever is that they’ve got these few really bad variants which have been pushed to high frequency, then you could potentially just fix that. (Doubtless there are major logistical difficulties, but it seems a nice idea to think possible for at least a few seconds).
That also has implications for animal breeds that are heavily inbred (breeds of dogs), where you can just locate the bad disease causing variants that have become fixed in a breed and eliminate them in the germline.
Possibly also for some heavily inbred human populations? (South Asia? Why marry outside the jati when we can just find whatever the jatis specific common large effect bad mutations are and engineer them out? 😉 ).
Another interesting thing for animal breeding is that, variants of very large phenotype effect size (which tend to be rare in humans except for skin colour?) seem fairly common in domesticated breeding experiments, where we have dogs where https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3847769/ – “Derived variants at six genes explain nearly half of size reduction in dog breeds”. So perhaps you could just take your Mexican chi-wah-wow and swap in your Great Dane large size variant, and then its easier to “explore” some of the entire trait space in animal breeding.
…
Slightly related to the topic of how normal relatively cheap genome sequencing became in 2010s: Lazaridis retweeted this BGI paper -https://www.biorxiv.org/content/10.1101/2020.12.10.418962v1 with a claim of about 100x drop in WGS cost. (“These improvements drop the turn-around time from days to twelve hours and the cost for whole genome sequencing (WGS) from about $1000 to $15, as well as increase data production by several orders of magnitude.”)
I’m not really technically able to assess this, but (besides many other major implications) so on ancient dna, having watched some video presentations by David Reich recently, he kind of emphasized that the advantage of capture/enrichment methods is reduction in cost, where there is a large amount of non-human dna (see – https://youtu.be/xF18mYe5WoA?t=198), and that relates to how the reduction in cost per genome kind of stalled out around certainly 2015.
However if you can drop the cost by 100x (or almost as much), then could we end up in a situation where, to avoid these questions about any slight biases or attractions induced by capture methods, its cheap enough that you can just use these massive shotgun sequencing methods instead and not worry about trying to design around bias?
I just listened to your conversation with Dilbert about the genetics of COVID-19 risk — very cool! But when I Googled “Traitwell” the description that came up in the list for traitwell.com looked something like this:
Maybe the site is just really new and this is temporary, but it could be that something is misconfigured, and if so you need to get it fixed, because this looks really sketchy.
Have you heard of the book The Revolutionary Phenotype? The author offers a new theory relevant to the post-human stuff.
https://www.amazon.com/Revolutionary-Phenotype-amazing-story-begins/dp/1729861563