For example, we need calcium. How much? Who really knows? There is variability of calcium in food (depending upon how it was grown: soil nutrients, water, sun, weather, etc.). There is variability in digestion, absorption, metabolism, transport, storage, excretion ... depending on disease states, time of day of ingestion, gender, age, physical activity, other foods eaten. There are a ton of inputs for the giant equation that makes up, that tries to solve, just how much calcium one needs. And those variables on the left side of the equation aren't static either! Our needs are always fluctuating. Also, when you tweak one variable it automatically tweaks another. So, it isn't just a matter of solving the equation, for the equation is never still. It is perpetually in motion and uncertain.
There's been a push-back against reductionist research in nutrition science in recent years. One place you see this push-back is in studies that interrogate dietary patterns instead of dietary nutrients. Nutrients, as described above, don't operate in a vacuum. They operate relative to other nutrients, e.g. carbohydrates act differently in a diet that has a lot of fat in it vs. a diet with little fat. Or you might say that fat acts differently in a diet that has a lot of carbohydrate in it, vs. a diet with little carbohydrate. I've written about dietary pattern analysis previously. Here's one post: High-Fat High-Sugar Dietary Pattern ("Meat & Potatoes") Linked To Colon Cancer, Diabetes.