On Tue, 29 Jan 2002 21:38:13 +1100, Richard Archer <[log in to unmask]>
wrote:

>As an exercise in SQL programming, I have also compiled a similar chart
>for all n-3 and n-6 fats, including DHA, DPA, EPA, 18:4 n-3 and 20:4
>n-6.
>
>This chart is much more meaningful than the previous one since it
>considers all EFA's including EPA, DPA and DHA.

Great pages Richard.
I'm just set up with my USDA Database (still sr13) and am now working on
meaningfull output, as time permits.

Let me make a few remarks on w-3 and the UDSA database listings.

The ratio only list has one drawback:
It lists all the very best w-3 sources, like Salmon/Cold water fishx,
but as you see on the amounts, the total amount often is often very small.
With Ratios like 11 or 17 you'd probably en up in a w-6 defficiency,
if you ate only Salmon fat. 0.25 is still a good ratio, the bigger ones
would be usefull to equal out other bad fats, and then it's the amount which
counts.

The main problem leading to the unfavourable ratios Cordain points to are:
Fats which are very high in w-6 (very big w-6 amounts) and little w-3
(example: sunflower oil, safflor etc.).
They don't appear in your listing because they have such a low ratio.
After all a ratio of 1:2 to 1:4 is still good - that's 0.25 in your list.

I think we have to take care for the amounts (like you did in
http://www.juggernaut.com.au/food/omega3.html )

Did you consider to weight EPA/DHA higher?
In one aspect (prostaglandin production) these long chain items
are more effectfull. If the conversion rate from LNA to EPA is 1:10 then a
weighting of 10 for EPA might be appropiate.

Suggestion so far: discard very low amount items and multiply EPA and DHA
times 10. Include Ratios down to 1:8 or so.

I took flax as a test-item. It's 50% pure w-3 (LNA) but appears not very
high in both lists. Aactually UDSA doesn't list flax oil, it only lists flax
seed. Because of it's huge w-3 amounts (at  low price) it
would be the only way to deal with such fats as sunflower.

Next problem:
If you want to equal out a moderately bad fat source like beef,
you see it's ratio is not bad (1:1 to 1:2) but the amounts are very small.
Particularly compared to it's MUFA plus SFA, with both can interfere with
total EFA chemistry (by competition).
One study Erasmus (and W.Price) quote, tells about a ratio of 15:69
of EFA to MUFA+SFA beeing the critical borderline to "suppress EFA
activity".
Suggestion for this: to build a factor of PUFA/(MUFA+SFA).
Then multiply the ratio so far by this - this will take the SFA disturbances
into account.

I also tried to discard non-natural sources, how did you do this?
I added a table with a paleo bit, which I have entered to contain only
natural and unprocessed food items (plus a few usual like beef).

Would you care to share your SQL code?

regards

Amadeus.