Growing Veggies packed w/ Flavor & Nutrition


Hello, My name is Makena & I am new to the OpenAg Forum.

I am most curious about the breadth of vegetables, herbs, fruits, etc that you can grow in a PFC. This article helped answer that question and thought I would share it! Continuing the discussion from What can and what can't be grown?:

My next question: Are there limits to what you can grow that actually have a good flavor and nutrient profile? As an Registered Dietitian, I want to be able to recommend PFC veggie growing to my patients but only if they taste good & are packed with nutrients!

With the help of the research team at @ATBFarms, we are going to try to answer these practical questions as all of you are too! How exciting!


@PFCDietitian Some folks here will probably disagree, but I don’t think this project is a good fit for recommending to your patients at this time. It’s not so much a matter of what’s possible as of what’s reasonable and cost effective.

The current state of this project is that people are enthusiastic about future possibilities, experimenting with a variety of prototype designs, and sharing ideas. [edit: People are talking about possible ways of collecting and sharing data, but I can’t point you to a study on nutritional profiles from crops grown in food computers because, as far as I know, nobody has published one.] If you are interested in helping to develop things that your patients might be able to use some number of years down the road, you’ll probably find people here to collaborate with.

If you just want to grow some food, Caleb has spoken well of AeroGardens and DIY systems that you can learn how to build on youtube.

[edit: Here’s another post from Caleb earlier this year where he explained the general state of vertical farming and what kind of data OpenAg wants to collect]


@openag Could you speak to the current status of flavor and nutrition testing for crops grown in PFCs?

Perhaps I’m overly biased towards thinking the progress is still in early stages. Besides what Caleb and others have publicly written about or posted on github, I’m unfamiliar with what’s going on at your labs in Boston.


@PFCDietitian So glad you started this thread. I will direct conversation on this topic here. I have a couple of questions for you:

  1. How can we test our produce for nutritional content? Is there a “DIY” method that will get us in the ballpark?
  2. If you could grow any plant in a PFC what would it be?
  3. What to you would be the most nutritional crops for us to focus on that also would be tasty to kids?

I will say right now that CEA offers the opportunity to manipulate flavor and nutrition in most crops. The possibilities really are endless for soil-less indoor production, the limitations come really from the economics of the operation. That being said, the biggest companies out there right now argue their produce to be the most nutrition/flavorful available:

This article by NASA outlines some of the leafy greens most suited to PFC style production (they give excellent detail about their rationale of crop selection based on nutrients not found in processed foods):

Watch this talk of Caleb’s. It will start at 9:00 when he goes into the full recipe’s (it’s worth watching all of it you haven’t seen the TED talk):

Here’s an article by OpenAg researcher on how they’re using data to increase flavor:

@wsnook you should check out the flavor data that’s on GitHub. I’m fairly certain it is the same data that is referenced in the above article:

Here’s a couple of other threads that are relevant:

How much control is possible over the phenotypic results of plants
Complete design of the personal food computer (and possible alternatives of some components)
Open source project proposal - call for contributers
Libraries & Food Computers: PLIX Build (Public Library Innovation Exchange at MIT Media Lab)

Peter, thanks for the link to Arielle’s article on Medium–I’d forgotten about that one. It’s good!

I’m familiar with the repo you linked to with all the csv files about gas chromatography and mass spectrometry on basil. I didn’t mention it specifically because there isn’t any explanation of equipment (PFCv1? PFCv2? food server?), procedures, or how to interpret the results. I don’t know enough about organic chemistry to understand what any of those numbers mean at the level of how that basil compares to any other basil in taste and nutrition. Do you? Can you explain what that data means?


Thank you so much @PFCDietitian for starting such a great inquiry into two relevant aspects of cultivated veggies, fruits, and herbs: flavor profile and nutrient profile. Nutrient profiles are quite complex arrays to generate and generally involve quite sophisticated and expensive research-grade instrumentation. Here’s some background on one of our key research initiatives: One question that we at @ATBFarms are quite interested in exploring in the development of a hypothesis is: What specific climate parameters can be correlated with either an increase or decrease in antioxidant production in a specific cultivar grown under CEA and hydroponic conditions? We are now in the process of designing an experiment to test antioxidant activity arising in select leafy greens. We found an assay that we can use in measuring the antioxidant capacity in samples extracted from the plants we grow. The idea is to eventually generate a database containing a large number of samples taken over no less than 1 year’s worth of trials. One of our Research Associates, @PFCDietitian, is a competent, skilled, and certified dietician / nutritionist, and she will be working with other members of our team in the development of our antioxidant tests for our “nutrient profile” research that’s being developed. Members of our team all agree that the PFC would be a great chamber for us to test our hypothesis. Of course, our research is totally contingent upon securing funding, and we’re presently working on procuring it.

There’s a big “if” dangling in the balance, yet once we begin to generate data for our antioxidant profile, then we’d love to explore in-depth discussions with others who share our enthusiasm for analyzing the nutrient content of food grown in the PFC. We find that there is quite a bit of alignment with our research and others who are trying to answer an important 2-part question: “How do people know that what they’re eating is really good for them, and do they know the amount(s) of nutrients they’re getting?” We all know that there are no labels on fruits or vegetables that tell you anything about the conditions under which they were grown or how much antioxidant is present. We also can’t forget that any labels that are placed on food for consumers are highly regulated by the FDA and USDA. So, how do we circumvent the label issue? Well, I have no clear answer for that one, but perhaps we can help open the door of nutritive transparency through independent and collaborative food research.


Some thoughts, those of us that have been looking at better understanding soil for many years have realized more and more how complicated it is to provide and constantly adjust complete plant nutrition. From the early focus on NPK to the understanding of more than 40 needed elements. These tending to be adequately available naturally in the sand, silt, clay but limited plant available without the proper microbial community. And that plants constantly naturally regulate the profile of nutrients with the cooperation of its micro-organism community. So instead of thinking about feeding the plant we look to understanding and stewardship of the soil life community and the working relationship with the plant. Then we can learn if the resulting produce is more inline with the complex needs of human nutrition. From what we learn from the relationship I can see it may be possible in the future to detect and understand the exudate makeup put out by the plant to trigger the release of the appropriate elements, in a closed system like the plant computer.
I like the thoughts of a home food computer for the present and into the future. And can see where it can certainly compete nutritionally with basic fertilizer food grown in lifeless dirt or some present forms of hydroponics produce presently available. How it will match up into the future should also be considered all along the way.


@ATBFarms That sounds cool. One thing you should understand is that the PFC v1 & v2 designs are still early stage prototypes, so repeatability of growth conditions is a challenge. For example, if you look through the forum, you’ll see that people often have trouble with sourcing lights, sourcing chillers, and operating the peristaltic pumps.

There have been a couple posts recently from the MIT folks about new designs they’re working on, but so far I think those designs are still private [edit: nope]. See Greetings Community - Lighting PCB (comment 15).

[edit: Github links for work in progress on a new designs:



This is exactly the kinds of questions we need to be asking. I would be very interested to hear what opportunities or publications @arielle may be aware of with regards to increasing antioxidant content in leafy greens. You should take a look at this paper by NASA called: Selection of Leafy Green Vegetable Varieties for a Pick-and Eat Diet Supplement on ISS @wsnook @ATBFarms @PFCDietitian @sichen @webbhm It is extremely helpful for establishing a baseline of crops to consider and provides an excellent weighting system on page 12.

Approximately 20 mg of the dried sample was added to the ASE350 (Dionex) extraction cell. Sample was extracted in Acetone: Water: Acetic Acid (AWA) at a ratio of 70%: 29.5%: 0.5% to a final volume of 11.7 mL. Aliquots of 0.2 mL of extract were tested for antioxidant capacity using the ORACFL assay. Dilutions of the extracted leafy green samples were prepared in a 75 mM phosphate buffer. Aliquots of 20 µL of the diluted sample were placed in a 96 well transparent flat bottom microplate, along with appropriately diluted Trolox standards, under subdued light. A “forward-then-reverse” pattern was adopted to place samples in the 96-well plate to account for possible positional effects. Additionally, the edge wells were not used for standards or samples due to possible “plate effects” which can result in significant measurement variability as compared to other parts of the plate. All samples were analyzed in triplicate (analytical replicates). The BioTek Synergy Hybrid plate reader automatically added 200 µL of florescein to each well, followed by 20 µL of the peroxyl radical generator (AAPH [2’2’azobis(2-amidinopropane)]) which then reacts with the fluorescent probe, fluorescein, to quench the fluorescence signal. The assay is monitored on the reader every few minutes for 2 h, and the presence of antioxidants extends the time required for the fluorescence decay to reach an endpoint. The ORACFL value is calculated from the area under the decay curve, and is reported as µmol Trolox (a Vitamin E analog) equivalents (TE) over a range of concentrations. International Conference on Environmental Systems 15 ORACFL has been established as a useful assay for the in vitro investigation of antioxidant capacity because the peroxyl radical produced in this a say is one of the foremost radicals present in the human body.

Greetings Community - Lighting PCB

Greetings @Webb.Peter, It’s great to see that you find resonance with the topic being discussed with respect to antioxidant capacity. We’re currently looking into a similar ORAC assay, and here’s a good paper [ORAC_App_Note.pdf]
on the experiment that has been tested by BioTek. In essence, the greater the disappearance of the fluorescein, the less the antioxidant activity (A.O. capacity). We thought this seemed to be the best assay to currently use, and the software they provide with the Synergy Multimode Microplate Reader will actually generate a standard curve, and you can see how well your sample “fits the curve”. We’ve already been in touch with a sales representative at BioTek, and the Synergy instrument is NOT cheap, yet we’re confident it is the appropriate instrument we’ll need to run our antioxidant experiments. We’re still waiting on some funding to come through before we can move on to next steps, and so we’re keeping fingers and toes crossed in hopes that we get the green light.


@ATBFarms I’m curious to hear how your program is progressing.

I thought you might appreciate this article from Plenty. One thing they certainly have is impressive market research and data regarding existing food supply chain waste and nutrient loss:

“Fruits and vegetables begin to lose their nutrient content as soon as they are picked. Spinach, for example, can lose 90 percent of its vitamin C within 24 hours of harvest, peas lose 55 percent, and other vegetables lose anywhere between 15 and 55 percent. The same degradation occurs with B vitamins, phytochemicals and other valuable nutrients and compounds; researchers use vitamin C as a marker because it’s easy to measure. Buying local is the only way of ensuring that the produce you put on the table is as nutritious as it can be.”

Do you (or anyone else here) know how to measure vitamin C cheap/easily? I’d love to include that or any other nutritional measurements as part of our MARSfarm curriculum.


Hello Peter,

I hope you’re doing well, and thank you so much for sharing the blog article from plenty. Our projects are proceeding along as planned, and we’re really excited to continue our energy monitoring and analysis component, along with some really exciting reconfigurations in the first phase. In the 2nd phase, we plan to expand our model, and then begin our initial design for the container farm. We’re really stoked to be in this position currently!

With respect to testing for Vitamin C content, the best way to do it is through a university (or perhaps private industry) Food Composition Lab. Bob Durst runs the OSU (Oregon State University) lab at the Linus Pauling Institute here in Corvallis, OR. I’m not sure what their pricing structure is currently, but perhaps they may be a good fit for what you want to have done. I suggest having a good assay done on vitamin C content, and try not to conduct a “do it yourself” test, unless you’ve done these assays quite a bit.

Be well,



@ATBFarms I appreciate the information.

Do you have any plans to publish any of the information or share designs? I would love to collaborate around container farm or perhaps targeted recipe development based upon post-harvest nutrient analysis.

I understand that an assay would provide the most accurate data, what I’m trying to understand is more the value of data obtained from a method like the one used here for Vitamin C:


Hello @Webb.Peter,
I apologize for the extreme delay in responding to your previous post - life has been replete with plenty to do. In reply to your question regarding publication, we don’t currently have that planned as of yet. We’re still working on vertical farming design configurations, testing hardware and software integration (and such), and it’s still taking quite a bit of work getting things off the ground. We had to put off conducting the antioxidant assays until we furthered our other work…we have learned much about not putting the cart before the horse though, mostly via trial by fire.

I hope to provide some more salient updates when we have a more complete system, and yes, we are working towards a system that can be integrated into a compact indoor vertical space (such as a shipping container). There’s still much to do, and the process is quite labor intensive (and at times somewhat costly). Be that as it may, we love the progress we’re making, and the PFC has been a great test apparatus for our work. I look forward to more updates down the road.