Jump to content

Database Sorting By Ratio


merliot02

Recommended Posts

Hello,

 

I need some suggestions on where to look for the following:

 

This is the issue that I have:

 

I need to sort the USDA database by ratio.  For example, using Nutrition Data, I can sort all foods in the Nuts and Seeds category in this manner: Highest in Fat AND Lowest in protein - as an example. http://nutritiondata.self.com/foods-012015000000078000000-1w.html?

 

But the foods ranked from #1 to #126 in this list does not coincide with the ranking that would obtain were the foods to be ranked by the RATIO of fat to protein.  The one-number ratio of fat to protein (2X/5X/10X) is what I need. [The carb to protein ratio also is of interest to me.  Dr. Wurtman of MIT, maybe he's retired now, wrote a paper years ago about how managing the carb to protein ratio could be (was?) important in managing Parkinson's or Alzheimers (don't remember which).  If anyone is interested, I'll try and locate a link to the paper and post it here.  The ratio was 5:1, by the way.] 

 

So I don't know where to go to be able to do a ranking of this sort.  No web-based database that I know of enables such a capability.

 





 

Steve Cushing
Link to comment
Share on other sites

This isn't web-based information, like what's available at Nutrition Data, which is what I'm looking for, but rather a software program. http://www.walford.com/software.htm

 

In the link, it says: 

"Dr. Roy Walford's Interactive Diet Planner - 

Now available with sustaining membership to the Calorie Restriction Society."

 

This is a forum on the Calorie Restriction Society website, so I presume someone can answer definitively whether this software program can do algorithmic calculations.  It may very well, but I sure would like to find this capability in a web-based format.
Link to comment
Share on other sites

  • 2 weeks later...

DWIDP hasn't been updated in a very long time, and was pretty crude by today's standard; I doubt it can be run on contemporary machines, and don't remember it having this capability.

 

I agree that this would be a great feature. I have often used NutritionData's nutrient search tool in the past for multi-parameter nutrient searches (CRON-O-Meter's is just for nutrient density), and have been baffled by the results when you just try eg. "high in fat" "low in protein" "per Calorie" (or vice-versa) -- which should, in principle, do roughly what you want it to do.

 

Sorry I couldn't be more help!

Link to comment
Share on other sites

  • 3 weeks later...

I got a moment!

 

On my drive, in the folder "macroratios" are two pdf files with ranked data for both fat to protein and carb to protein.

 

The readme explains what I did and if it isn't what you were looking for Steve, it would be trivial to rerun these analyses so let me know. Same with the format if pdf isn't what you're after.

 

https://drive.google.com/folderview?id=0B10kLHx60XW8SWRucEQtUzF3WkU&usp=sharing

 

What really interests me is whether it is possible to capture the notion of "ideal foods for calorie restriction" into a mathematical formula. I've been playing with this idea for a while. For example, if you define a "good" nutrient as one that it is hard to overdose on and a "bad" one as those one would like to avoid completely (trans fat, DHA?) it is fairly simple to perform a ranking of foods where calories are inversely proportional to good minus bad nutrients. It would be nice to add in other information relating to glycemic impact, and possibly antinutrients like phytates etc.. These macroratios might well figure in such a model. What becomes difficult is weighting the importance of these various factors, in a principled way, when you put them all together. If anyone has any ideas about this or has an analysis that they'd like to see performed on the usda database, I'd be happy to run it.

 

Eugene

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...