How much control is possible over the phenotypic results of plants



I’m working on a design concept inspired by the Food Computer and a key feature is to grow for different phenotypic results on the same type of seeds - say, one type of Basil. I was wondering how much control is possible over the phenotypic results at this point in time and where you see it going in the future with more data being aggregated?

How big of a difference can you get in terms of say taste or color by tweaking the environment? Is it in the realm of less taste vs more taste or can plants become more salty or sour based on how you combine nutrients and the growing environment?

Any input on what has already been done or you see as possible in the future, would be greatly appreciated.

Thanks in advance.


I don’t think the question is as simple as you present it, as I don’t think we know that much about taste influences, and less about the differences of various plants. How much of taste is genetic and how much is controlled by environment? Genetics is a big factor, and one that still needs a lot of exploration, and the genetics control the ‘plasticity’ of expression - how much phenotypic variation is possible for a particular plant variety. You can do all of the environmental manipulation you want and you will not get a habanero pepper to taste like a Fuji apple.
It is hard to call flavor a ‘fixed’ property of a plant, but it is something more dynamic. As a fruit ripens, there are continuous changes in the amount of sugars, as well as the other chemicals that give the flavor. The question is more of: given the genetic potential range of flavors, what environmental influences at time X, produce what flavor at time Y.


It depends on how you qualify the scope of your question.

If you ask, “How much control is possible with knowledge and equipment from the fields of horticulture and controlled environment agriculture?”, the answer is “Probably a lot, but It’s complicated.” You might want to try reviewing the literature. [edit: see this thread, Technical books, publications, and links on hydroponics & CEA, for starting points]

If you ask, “How much control is possible by building and operating a OpenAg PFC v2.1 food computer, an MVP food computer, or one of the OpenAg food servers at MIT?”, the answer is, “That’s an open question.” People have grown some crops, and some data has been collected. But, no one has publicly documented food computer or food server experiments designed to give statistically significant, reproducible answers about phenotype expression. That’s more of an aspirational goal people are talking about as opposed to a current capability.


I don’t know. But like others have said it probably depends. If you are talking about a whole plant like lettuce then something like controllable high intensity UV Leds could affect the amount of purple/red color (anthocyanins) if that variety of plant has anthocyanin capable genetics (some do not). Anthocyanins can affect flavor. Some people think they make things taste more bitter, some people think they make things taste better. Depends on the genetics of your taste buds i guess. I for example am allergic and hate the taste of Cilantro.

If you are talking about flavor and phenotype of tomatoes then something like UV exposure, temperature, AND possibly Infrared exposure (heat) affects the types and accumulation of sugars in the fruit. I know from growing things and my plant breeding network that plants bred for one specific climate (like potatoes or squash) when grown in someone elses climate or more particularly soil with different nutrients have different flavors. One thing grown by you may taste different from the same exact variety grown by someone else in different soil.


This is such a good point, flavor is arbitrary. There never will be one perfect variety of lettuce, because genetics impacts taste buds and so do things like culture and background.

Do you do any analysis of your soil? Something I’m very interested in is “translating” lessons learned from soil growth into recommendations for hydroponic/aeroponic recipes.

Might check out this thread too:


No i havn’t. i really should try. I have a soil test kit, but it was hard to figure out. maybe i’ll try again. One problem though is the soil in my yard varies so much. There are places where not much grows like a dessert (only the peach tree and some short sage plants grow there) and other spots where manure was brought in years ago to improve the soil where the nutrients are fairly good. And everything in between all in the same yard haha.

I read that paper! pretty cool stuff! Haven’t watched the ted talk yet. will later.

Do you know where that linked photo goes to? I couldn’t find any information about that science fair looking project with HIgh UV on the left and no UV lettuce on the right. very interested in that.


Picture is stolen from this talk,

Dr. Cheri Kubota out of University of Arizona has an incredible series of talks on Controlled Environment Agriculture (CEA). Her work is mostly focused on berries and other small fruiting crops.

Here’s another one of my favorite images from Kevin Folta’s work @ericeisele @Drew @BioLumo:


Sorry late at night and i can’t sleep. I’d like to talk about this a little more. +1 i pretty much agree with everything webbhm said above.

I think what your question is really dealing with is genotype vs phenotype and how much environmental changes can influence that expression. I will give my current thoughts on this.

  1. Based off of my experience with vegetable breeding and observation and growing unusual varieties of plants and vegtables these last few years i think genetics or genotype plays a HUGE role. Much larger than most people probably realize. For me in a regular old back garden starting out with good genetics or good seed is huge. Sure i can take any random mal-adapted variety of “plant z” and grow it enough times over and over and save those seeds until enough epigenetic changes or random mutations or natural heterosis and differentiation occurs that the original “plant z” does well for me in my particular environment. But a better approach would be to start with varieties that already have a few legs up in this area. Something similar might need to be done when selecting varieties of vegetables or fruits for artificial cultivation in something like a PFC. Sure i’m sure you could grow any ol’ type of “plant z”, but “variety y” or “variety x” might do better from the get go.

  2. Like i mentioned before i think one should not forget that taste can be arbitrary to the individual as we all might have different genetics ourselves and different taste buds or olfactory senses, etc. But that does not mean that certain phenotype can not be influenced or changed by the environment. Take for example a variety of Pea that i am currently working with. The Pea variety in question is called ‘Mighty Midget’. It is unusual in the fact that not only is it a dwarf pea, but it is an “extra dwarf” or what i like to call it a “super dwarf”. On average it is said to grow about 6" tall. For me it generally grows 1"-4" tall. I don’t know if peas would grow well in containers or a PFC like environment, but since it is so short it is a good variety to experiment with or use in breeding. But anyway, this last season i planted it in a large group of other pea varieties. These other peas are super tall, many of them reaching 6-7ft high and blocking out a lot of the light. When i went looking for this pea later in the season under all the growth of the other peas i found that this plant had grown very tall compared to normal. It looked like a normal average pea variety as it was probably 2ft tall.

  3. Taste probably can be influenced depending on the plant species, variety, and the environmental input. Some plants that survive in salty environments sweat or store salt through their leaves. It might be possible to water stress a plant to become more salty, even in the fruit like a tomato. But this would require a non-hydrponic setup (i.e. soil). I’ve heard that certain varieties of domestic tomatoes have naturally more salty fruits than others because of genetics, but i wouldn’t know for sure. Domestic tomatoes despite seemingly having a wide genetic base with various shapes, sizes, colors, and tastes are actually quite narrow genetically suffering from several bottleneck events during multiple domestication periods. That is why many tomato varieties are susceptible to the same diseases. It is becoming a bigger problem as time goes on.

  4. Some things can NOT be influenced environmentally without certain genetics being in place. It is true that plats that produce red or purple pigments (anthocyanins) are influenced by temperature and UV exposure. But lets take Purple pigmented Peas or Corn (or even pigeons) for example. purple podded peas require a “master switch” gene called [A] to turn on purple color in the first place. The same principle is the same in Corn and even the color of feathers in Pidgeons. If that gene is recessive [a] then even in UV and temps are satisfied the plant will have no purple color (or the pidgeons will be white lacking color). Hence white domestic pigeons or white laboratory mice.

Now, let’s say that a certain Pea plant has a dominant gene [A] master switch color gene. That means that yes it can use color. But it needs other genes which are controlled by [A] or are epistatic to be there as well. In many peas they will show purple axil splotches around the base of their leaves if gene [A] is present. If they have other color genes they can have a wide range of other phenotypic traits such as purple pods (requires two co-dominant genes + [A]), incomplete or partially-purple pods, purple or red seeds, etc. In the pidgeons this can be anywhere from blue to white with black tail feathers, to brown, to black. Same with mice.

So thus, even with the most elaborate environmental tinkering you are still not going to make a purple podded pea appear out of thin air from a green podded pea. Or apply the principle to your plant of choice like Basil.

  1. like webhm said, “the genetic potential range of flavors” is what comes first. I am working with some [Solanum pennellii x domestic tomato] hybrids. Solanum pennellii is a wild tomato that has small green fruits, but it has all kinds of alkaloids that could affect the flavor of future tomato lines. Some of these alkaloids probably taste bad and are unpalatable. But others have the potential to enhance or change other tomato flavors for good (or even make them more healthy with more amino acids or something). One plant breeding collaborator of mine is doing even more wild tomato crosses to domestic tomatoes and backcrosses and multi-species crosses. He swears that some of his new tomato lines taste fruity like pineapples or other strange new flavors not commonly associated with tomatoes. So who knows. But you can’t breed or grow a pineapple tasting tomato without first having a wide genetic potential of genes to draw on. CRISPR is a cool new technology, but you can only use it to manipulate what is already there. at least as far as i know. I really don’t want to introduce CRISPR to this discussion. Just mentioning it briefly. Hope i don’t regret that later on.

idk. i guess that’s all my thoughts right now. You might be able to change the environment to make a plant produce bigger leaves or larger roots or something small like that. Hope that answers your question in way more depth than you even wanted to know about. :wink:


Any thoughts on how to define (quantify?) flavors. I know that the wine industry has done some work in this area, and even have ‘test kits’ of the standard flavors. There is also the Scoville scale of hotness for peppers. It is hard to get quantifiable research when we don’t have a consistent way of describing the results.
As a start there is always the binary “Like”/“Dislike”, or a 1 to 10 preference scale. These are highly subjective, but better than nothing.


good question. I have no idea. Though i am highly interested in seeing others answers for this. I will have to do some brainstorming. I’m sure there will be something we can come up with.

With Watermelon fruits i know you can use a Sugar Refractometer and measure Brix. I have never used one but i’ve been meaning to get one. The Brix gives you a rough idea of sweetness and is probably pretty good. But i know from talking with other people that Brix itself is not the whole story. For example one guy told me something like that one watermelon he tested had a Brix of 14 and was a red fleshed watermelon. Pretty sweet. But he said he had harvested a yellow fleshed watermelon that had a Brix of only 10. But oddly out of the two he preferred the yellow fleshed and thought it actually tasted sweeter to his tongue. Interesting huh? I certainly think so. This of course means you actually have to cut the fruit open it to test ans taste it. I don’t know which sugars the refrectometer is actually testing for but there are lots of different plant sugars which all may have an effect on taste that i don’t think can all be tested for with a refractometer. And this is just sugar. What about all the other things like carotenoids or anthocyanins which may also affect flavor?

Another idea i had heard about, but which i don’t think ever got finished being developed was a sensor that could detect ripe watermelon from outside the fruit. Highly worth having had it worked flawlessly, but i think they were having some problems with it. Still highly interesting non-the-less.