Are you sure about that? Wikipedia’s Nutrigenomics page seems to reference a lot of articles on documented gene-nutrition interactions, incuding the effects of nutrients on genetic expression.
This system would need to be based off an awful lot of data to be producing such specific prescriptions
I don’t think SWAMI actually needs that many pieces of data to make strong recommendations; it claims to be using only 225 of the nutrients or substances found in 800 foods as a basis for its suggestions.
As I understand it, it’s essentially doing something like, “people with this set of genes tend to have these problems; these nutrients tend to help with that kind of problem, these others make it worse—so rate the foods containing those nutrients up or down accordingly...” and then it computes a total score for each food, and uses various cutoff levels to rank the food as “good”, “bad”, or “meh”. ;-)
IOW, it’s not using a massive array of studies on individual foods’ effects, but rather, a scoring system based on known nutrient-genome-health correlations. And statistical prediction rules can easily outperfrom human experts, so it shouldn’t be especially surprising that you could get some pretty good results out of less than “an awful lot of data”.
It would be an excellent thing to be working towards but right now does not sound credible
On my epistemically rational side, I would certainly like to see more references myself. D’Adamo’s book and software describes many kinds of “this does this to that and is related to gene XYZ-123” things that cause my brain to go “[citation needed]”—i.e., I would really like to have a better idea of what his epistemology for all this stuff is, besides, “we studied it in my lab”.
On the other hand, my instrumentally rational side has been happy enough with the results from following the book’s recommendations so far, to be willing to buy the full kit. The interesting question will be whether I can lose more than the typical “20 pounds and then start regaining” that happens when people switch to new diets, and that will take a bit longer to determine.
(OTOH, I’m already about 20 pounds down from my last major dietary change about 8 months ago… so perhaps any further weight loss will be a good sign.)
Are you sure about that? Wikipedia’s Nutrigenomics page seems to reference a lot of articles on documented gene-nutrition interactions, incuding the effects of nutrients on genetic expression.
I don’t think SWAMI actually needs that many pieces of data to make strong recommendations; it claims to be using only 225 of the nutrients or substances found in 800 foods as a basis for its suggestions.
As I understand it, it’s essentially doing something like, “people with this set of genes tend to have these problems; these nutrients tend to help with that kind of problem, these others make it worse—so rate the foods containing those nutrients up or down accordingly...” and then it computes a total score for each food, and uses various cutoff levels to rank the food as “good”, “bad”, or “meh”. ;-)
IOW, it’s not using a massive array of studies on individual foods’ effects, but rather, a scoring system based on known nutrient-genome-health correlations. And statistical prediction rules can easily outperfrom human experts, so it shouldn’t be especially surprising that you could get some pretty good results out of less than “an awful lot of data”.
On my epistemically rational side, I would certainly like to see more references myself. D’Adamo’s book and software describes many kinds of “this does this to that and is related to gene XYZ-123” things that cause my brain to go “[citation needed]”—i.e., I would really like to have a better idea of what his epistemology for all this stuff is, besides, “we studied it in my lab”.
On the other hand, my instrumentally rational side has been happy enough with the results from following the book’s recommendations so far, to be willing to buy the full kit. The interesting question will be whether I can lose more than the typical “20 pounds and then start regaining” that happens when people switch to new diets, and that will take a bit longer to determine.
(OTOH, I’m already about 20 pounds down from my last major dietary change about 8 months ago… so perhaps any further weight loss will be a good sign.)