Adrián Ferrero, co-founder of Biome Makers, discusses how DNA analysis of soil microbes can help growers better understand what’s happening underground.
Acres U.S.A. Can you tell us a little about your background and why you founded Biome Makers?
Adrián Ferrero. I’m a passionate entrepreneur. I’m originally from Spain, and I have a degree in economic science, and I started working to support corporations in innovation projects, especially trying to find funding resources — especially connected to the European Commission. I spent almost 10 years supporting companies in different sectors there.
One of my childhood friends, Alberto Acedo, is a Ph.D. in molecular biology, and he knew that I was supporting innovation initiatives and entrepreneurs. He had this idea of applying next-generation DNA sequencing for doctors for disease prediction, or predisposition to genetic diseases. So, we decided to team up and create our first company, which was the first diagnostic company mainly focused on hereditary cancer. We were the first to use saliva as the sample tissue to diagnose this predisposition to cancer, and that was very valuable for doctors.
We launched that company back in 2012, and we won a lot of recognition — from the European Commission and the Spanish government and from MIT Technology Review. That was good, but we realized it wasn’t enough. So, once that company was growing, we sold it and decided to apply the same concept to agriculture — personalized agriculture, just like personalized medicine.
The reason we decided to do this is that we’re both originally from the northwest of Spain, an area called the Spanish Midwest. There’s a lot of corn there; sugar beets and wheat are the other main crops. And at times we were not able to drink the tap water because there were so many chemicals in it. Instead of blaming the farmers for this, we realized they didn’t have some of the indicators that could help them manage what was happening in the field in a predictive and efficient way. So, that was the trigger for the whole biometrics idea.
We decided to start in California because we got an invitation from the Illumina accelerator program. They selected just three companies at a time, and we were the first non-American company invited to this program. In California, there are elements that, back in 2014, you couldn’t find in Europe — people willing to invest in this crazy idea of applying biotechnologies and computing science to agriculture. Those are three big challenges at the same time. But in California you have those industries, and agriculture is especially important here.
We launched Biome Makers in May 2015. It was just Alberto and me, and we started working in one single crop: wine grapes. We were able to close a seed round of investing, and today, we are working in more than 189 crops, with 80 people on the team, and with operations in 45 countries. Although, the reality is that we analyze samples from all over the world with the support of our lab partners in France and Guatemala, among others. We have the largest soil microbiome database for agriculture in the world.
Acres U.S.A. Acres U.S.A. readers are of course early adopters of this idea, that soil biology is important, but you must be very optimistic about the potential for this. It seems like it’s becoming a mainstream idea.
Ferrero. It is becoming mainstream, and we’re grateful to be involved in promoting the idea that the life of the soil is important. But we actually try not to focus on the individual microbes in the soil, because we realized that that was too scientific of an approach, and not everybody was willing to explore that depth of data. That was a learning curve. When we started releasing results on the types of microbes present, we received feedback from winegrowers that what was important to them was not which microbes were there, but what those microbes do — what functions they perform. People are starting to talk about the ecological services microbes provide.
This idea of soil functions was what helped us provide similar value to the whole spectrum of growers. We have a lot of operations in the Midwest, many of which are more traditional operations — growers who are yet to adopt regenerative agriculture practices. Although, it is clear to them that the cost of inputs are growing very quickly and that the profitability of their operations is going down, and they have to do something. They also see that the productivity — the natural fertility — of their soil is going down. They know that dependency on certain inputs is not good.
What is key for us is helping them understand the relevant pieces of the life of the soil. It’s not about having one microbe over another. Individual microbes are important, but it’s more like human societies — when you want to understand how a city works, you have to analyze it from a different point of view. You have to know how many police there are, and how many bakers, and doctors, and bankers. It’s the whole structure of the community that delivers all the services that you need. The soil is much the same. It’s an organism — a community of living beings — a system composed of different individuals that belong to different species but that interact with each other and have relationships.
So, we create simple indexes to measure these functions. The growers we work with — whether they are organic or conventional or are starting to introduce regenerative practices — at the end of the day, they all have a similar question, just with different concerns. The question is, “What do I do in my field?” We’ve identified three areas of interest that we think are valuable and as actionable as possible.
The first one is nutrition. Whether the grower is applying synthetic fertilizer or compost or compost tea or cover crops, or nothing, what is happening with fertility — with the nutrition of the field? We’re not talking about the nutrient content, though — we’re talking about the biological processes happening in this soil. We measure the mobilization of the macronutrients and the micronutrients — the good pathways, the negative pathways and also the neutral pathways. At the end of the day, what is important is the overall assessment of each one of the pathways for each one of the nutrients. If you’re an agronomist, you can go deeper and look at the details of this.
Acres U.S.A. Are you saying that if a plant tells its microbiome that it needs, say, molybdenum, that you’ve identified the specific bacteria or fungi that can produce molybdenum?
Ferrero. Yes, and let me get a little technical here. First, we look at the DNA of the organisms in the soil, and the technique we use allows us to collect data on all the bacteria and all the fungal species that are present in the soil. Many of them have not been named by humans yet, but we can do a taxonomic association. For example, we might not know if we have lions, house cats or tigers, but we know we have a feline species.
The second thing — and this is the most interesting piece of the technology — is the scalability that computing science now gives us. Every time we process a sample, our system learns. It starts to create an idea of what that soil looks like. We build references for each crop, based on the samples we analyze, so we know what is normal and what is unique, and then we connect that to other metadata. That uniqueness could be positive or not as positive for the farm.
Finally, we’re able to make a correlation between the microbes and the pathways. There are many scientific papers that discuss different microbes and their functions. Every day, researchers across the world analyze and publish the full genome sequence for specific bacteria and fungal species. So we have access to all the functions that they have. One of those functions might be connected to, say, nitrogen mobilization. But is it about just adding microbes according to the functions they have? Of course not. We have to keep in consideration that they’re in a community. Some of them are not going to be able to do anything or perform any function, or will do different functions, depending on the rest of the community. We have to study the community from a network point of view — their co-dependencies, co-exclusionary relationships, etc.
It’s just like how banks are able to predict your credit capability based on analyzing your social networks. They know how you behave — who you’re talking to, certain sports you like, if you have connection to certain risks. This is the same kind of technology, but in this case applied to understanding the full microbiome. We know which microbes are in the soil, we know the functions the different types of microbes perform, and we can understand the ecological relationships. And with that, we can identify and measure the different pathways.
Now, one thing that is so interesting in the microbiome is that it’s a natural biosensor. The moment you put a seed in the soil, the microbial community is going to adapt to that new situation. If there is a lot of water, the microbes are going to react to that. But there is a community called the core microbiome that remains stable across time and is singular per crop, per region. There are different elements that are going to impact the composition of this community, but it’s those core microbes who play the largest role in all the biological activities that are connected to plant growth or crop performance. So, for us, studying this community is so interesting. When you look at microbes, you can start predicting, for instance, the risk of disease.
To go back to the three areas of interest that we try to highlight to the grower — the first being nutrition — the second would be soil health. But soil health is a really broad concept. People always try to define it. So when we say soil health, we’re talking about lack of disease. Somebody is healthy when their body works well — they balance defensive mechanisms with risks. We’re exposed to pathogens all the time — which is great, because at the end of the day, we’re training our immune system. The soil is pretty much the same — we’re able to measure the disease risk.
This is one of the key results that we deliver, that our clients appreciate. Because it’s not about pathogen identification, but rather measuring the risk of carrying diseases, and also other parameters like stress and hormones and adaptation factors. Microbes play a really important role in this. They impact the strength of the plant and how it’s going to grow or how the crops will develop.
And the third dimension that we analyze is the biological quality of the soil. What is the microbial biodiversity? How are its functions compensated? Is the community structure solid, or does it have weaknesses? So, these are the three groups of results that we deliver for every test that we do.
Acres U.S.A. Can you walk through how the process works? The grower takes a soil sample in the field, and they send it to you, and then what happens?
Ferrero. The sampling protocol is actually quite similar to any other soil sample. In fact, many people do both at the same time, because our tests provide a layer of data that is unique, that complements other tests. The idea is that you do your traditional tests and you complement them with this biological dimension that was previously missing.
Once we receive the sample, we start our DNA sequencing workflow. We generate DNA data, then we process it. We do everything in-house — first the bioinformatics pipeline to identify the microbes, and then we start connecting the dots of the communities. And then we calculate and deliver 56 indexes that are grouped into the three areas that I mentioned. There are different levels, depending on how deep you want to go.
Customers receive a PDF report, and the artwork we use is on a scale from yellow — low — to blue — high, and normal, which is kind of gray. People ask why we don’t use red and green, and that’s because the numbers can’t be inherently good or bad. When we talk about biology, yes, we can assume that increased biodiversity is generally good. But there are certain situations, depending on what you’re looking for, or depending on your growing philosophy, in which certain populations of microbes could be either good or bad. Plus, it’s not that the highest biodiversity delivers the highest yield, but having a high biodiversity does deliver a high yield.
Acres U.S.A. Some people even talk about using biodiversity as a marker — as an alternative to carbon sequestration — for payments to farmers. Do you think that might become a reality someday? DNA testing would certainly be important for that.
Ferrero. Biodiversity is a very good metric, and very good indicator on what’s happening in the field. But we should not get obsessed with just increasing the number of species.
If you have a high number of species, but one of them makes up 90 percent of the population, then you have very diverse soil, but the reality is that that soil is not diverse, because it’s dominated by one single species. And you could have a different case where you have a lower number of species, but the abundances are more balanced. In that case, you assume that the biodiversity is lower, but the community is stronger.
Some of our clients, when we tell them they have low biodiversity, become really upset. But then when you go into the details, they can identify the problem and the action point. Sometimes their previous interventions mean that certain species are dominating the ecosystem, and that creates a lot of weaknesses. Diversity is a really good indicator of soil health, but if we look at it by itself, it could be misleading.
Acres U.S.A. Soil biology can be used to analyze the “terroirs” of different places, especially with wine. To what degree can you use the microbiological profile of a place even forensically? Is that possible? Or could you change the microbiology of one location to match another — if you really like the wine that comes from a certain vineyard, can you change the soil microbiome in your vineyard to match it?
Ferrero. The human microbiome actually has been a forensic indicator in a murder case. With our previous company, a hospital asked us to analyze the microbiome of a human tissue sample to see if there was a match on the microbiome fingerprint. And that was admissible evidence.
As for terroir, for sure there is a signature, especially when you look at the species level. When you analyze the taxonomy, there are differences and similarities. But with wine, while the origin is so important, the winemaking process makes a big difference as well. There are certain tools the vintner can use to boost or erase flavors. For large wineries, their challenge is to be consistent with their flavor, while boutique wineries or highend wineries appreciate uniqueness over time — the vintages are different.
So, yes, there is originality — there is a terroir, or a fingerprint, a characterization based on the soil microbiome. But if we look at this dimension from the functional point of view, then we can compare practices. And this is what is more interesting in agriculture right now. There are so many growers testing different practices and coming back to traditional practices that are not based on inputs, and understanding how these practices can be applied in different locations.
For example, if you compare the Brazilian rainforest with the Australian rainforest, they’re completely different from a taxonomic point of view — the soil species are completely different. But the biological processes that are happening there are exactly the same. So that’s what we want to know — is there something farmers growing corn in Brazil are doing that could be applied here, or in Australia, to increase soil health and at the same time maintain or boost productivity?
Acres U.S.A. Can you talk a bit about applying biological products that a grower purchases, as opposed to nurturing the biology that is already in the soil? People have very different opinions on buying “bugs in a jug” versus just inculcating what’s already there. What do you advocate primarily?
Ferrero. It’s great that we’re seeing this revolution now in the tools that are available in agriculture. There are all kinds of prebiotics and probiotics now for the field. With bugs in a jug, you might be feeding your natural population of microbes, or stimulating them in a certain way. And there are new tools and new technologies to come. Many companies are working really hard to deliver new solutions that are going to be more targeted.
The way we see it is that tools are tools, and you have to use them depending on your needs. If you have a headache, you may just be able to drink some water and go to sleep. You might take aspirin. But what you’re not going to do is start chemotherapy. I’m being very radical in my analogy, but what I’m saying is that we have to understand all these practices as tools. If you know how to stimulate your natural biology, that’s going to be cheaper for sure. But perhaps at the beginning, you have to stimulate the soil with external inputs.
We’ve seen different cases of people who have been able to increase their numbers, and restore the life of the soil, by themselves. This is a circular economy — they take advantage of the byproducts and residues from their farm and compost them, and it’s working. But some farms need offfarm inputs, at least at the beginning.
We work with input manufacturers to unveil the real effects their products have in the field, in order to increase their predictive power. The benefit of computing technologies is that we’re starting to validate that our system is able to predict what’s going to work best in a specific field. And the system doesn’t always recommend that we need to add inputs. But biologicals often work really well.
The products that are not working — what is traditionally called “snake oil” — are getting out of the game, because it’s becoming clear that they’re not producing a clear or characteristic effect. Or, what you see is that they are adding nutrients to a biological product. But that specific input compared to another input with the same composition, without the biological, is the same. In that case, we of course don’t recommend that product.
We work right now with over 80 input manufacturers; we’re an enabling technology. We’re able to produce data on what’s happening in the soil, at a functional level, a biological level, and anybody can have access to this technology. We’re really agnostic. But what we want for sure, and this is part of our DNA, is to help to restore the cycle, the natural fertility of the soil. This is one of the goals that we have as a company — to give the farming community more tools and more information to make more qualified decisions on what they do in the field.
Acres U.S.A. What about in-field testing? How much more can you get in the lab than what you can from an in-field test today, versus what you hope five or 10 years from now? Will we be able to do all these things in the field that you’re doing now in the lab?
Ferrero. I love that question because technology is evolving. There are already tests on portable devices where you can analyze DNA in the field. But there are still so many complexities; we aren’t there yet.
But what we’ve realized is that we don’t need to wait for that. As we learn and understand the dynamics of the soil, we’re able to develop functional reports without soil samples, with almost 90 percent accuracy compared to an actual sample. This is an experiment that we have validated, and it’s going to help democratize access to this layer of data.
Acres U.S.A. How do you do that without a soil sample — what are you testing?
Ferrero. We’re inferring. We infer what is going to be the status of the soil functions based on available data. This isn’t commercial yet, but we’ve validated it internally.
We know, for certain crops — corn in the U.S., for instance — based on GPS location, our computing systems are able to develop at complete functional profile with a very high level of confidence — close to something between 85 and 90 percent, which is really good. There are some human diseases that are being diagnosed with less confidence than the accuracy we have for predicting soil conditions. We just need to have enough reference data, updated every year. We’re currently working to make this accessible. We plan to release it by next year.
We saw that if we waited until the technology was ready to deliver in-field testing, we’d probably be waiting too long, and it would still be too expensive. But if we can decrease the cost, that will allow growers to make good decisions about their soil.
When we started Biome Makers, we had to build the data and learn from the data to really understand how to use it to deliver better results. We learned by producing the data. And the system is identifying the patterns. We integrate probably 100 to 180 different parameters per sample, and we correlate that with climate, weather and soil structure, which are publicly available. It’s the magic of computing data technology, and right now we’re living in a moment where this artificial intelligence that everybody’s talking about is accessible. There’s a lot of knowledge that we’ve integrated into our system that will allow us to produce these profiles.
Acres U.S.A. And that would be independent of the way a farm is being managed? Because you’re predicting the microbiological profile of a farm based on its location, correct?
Ferrero. Based on a sample, we can of course test the sample and know what’s happening with the microbes on the farm. This method that I’m describing is without a test. So what if the grower is doing something different than their neighbor? In that case, we need access to some additional data. And all large growers today are using integrated platforms and producing data about what they’re doing. Our system will receive that information on the practices those growers are using, and then our results — our inferences — will be more accurate. And, of course, if they want to go deeper into the details and get a 100 percent accurate picture, they can then do a soil sample. And someday we’ll have infield technologies to do this.
But our current turnaround time for soil samples is not very long. Our commitment is four to six weeks, but the current delivery is something between 10 and 15 days. So in two weeks you’re receiving your results, which is really quick.
Acres U.S.A. And what is the ballpark cost of the soil microbial DNA test?
Ferrero. A single test is $199. But what is more interesting is the impact and the range of the test. If you just look at the individual price, compared to physical chemical test — which is completely different — yes, ours is more expensive. But if you’re thinking about the impact it’s going to have — knowing that whatever you were going to apply is not going to get into the plant, because your nitrogen pathway or phosphorus pathway is blocked — then you realize that the real cost per sample is really low. The cost per acre for the test, depending on the crop, could be as low as $3 to $6.
Acres U.S.A. Definitely. Testing is almost always worthwhile. Even if it only influences a single management decision, you’re going to save or make that much money at least.
Wrapping up, what do you think the future is for regenerative agriculture?
Ferrero. I think that regenerative is not just about a new organic label. It’s a mindset change. It’s about improving farm operations in ways that are going to be good for the farm and for the rest of us. That’s the real goal. There are always going to be massive farms that will be driven by yield. But even those farms can benefit a lot from regenerative practices. And the real innovators are the ones who open their books and show what’s happening. We’re trying to help by providing metrics that are reliable and trustworthy — to measure what is happening beyond what we can see. What we see aboveground is clear; we’re helping growers “see” what’s happening underground.