#79 | Genetic Improvements of Cup Quality: Latest Developments | Hanna Neuschwander, Expo Lectures 2019

#79 | Genetic Improvements of Cup Quality: Latest Developments | Hanna Neuschwander, Expo Lectures 2019

This podcast is brought to you with support from

World Coffee Research is working to improve the genetic potential of coffee cup quality. To start this task, you first need to define the meaning of “quality.” WCR believes it is a market-driven definition and organized new protocols to assess quality from the perspective of the coffee industry. With experts and partners, WCR produced the coffee lexicon that allows an objective analytic description of coffee quality attributes. In their presentation, Hanna Neuschwander discussed these innovations and shared some lessons learned, including the latest results deriving from these methods and advances on the molecular basis of cup quality. In particular, Hanna highlights the burgeoning importance of some specific volatile compounds like Limonene. These results are not an end but a beginning; an opportunity for further studies to find out the molecular markers or genes related to cup quality in general. 

Hanna Neuschwander is the communications director for World Coffee Research, a research nonprofit that works to secure the future of coffee. She writes and speaks globally about coffee science and agriculture. Her writing and commentary has appeared in the Art of Eating, Time Magazine, CNN, BCC, Portland Monthly, and many others. She is the author of Left Coast Roast, a guidebook to coffee roasters on the west coast.

Special Thanks to Softengine Coffee One, Powered by SAP 

This episode of the Expo 2019 Lectures podcast is supported by Softengine Coffee One, Powered by SAP.  Built upon SAP’s business-leading Enterprise Resource Planning solution, Softengine Coffee One is designed specifically to quickly and easily take your small-to-medium coffee company working at any point along the coffee chain to the next level of success. Learn more about Softengine Coffee One at softengine.com, with special pricing available for SCA Members. Softengine: the most intelligent way to grow your business.

Related Links 
Table of Contents

0:00 Introduction
2:50 Why coffee farmers are in need of high yielding varieties that also have high cup qualities
14:40 Summary of an experiment that identifies desirable flavor markers in coffee plants before they produce cherries
31:20 Summary of the current genetics-by-environment interaction research efforts
44:10 Audience questions
52:20 Outro

Full Episode Transcript

0:00 Introduction

Heather Ward: Hello everyone! I’m Heather Ward, the SCA’s Senior Director of Content Strategy, and you’re listening to the SCA Podcast. Today’s episode is part of our Expo Lecture Series, dedicated to showcasing a curated selection of the extensive live lectures offered at our Specialty Coffee Expo. Check out the show notes for relevant links and a full transcript of today’s lecture.

This episode of the Expo 2019 Lectures podcast is supported by Softengine Coffee One, Powered by SAP.  Built upon SAP’s business-leading Enterprise Resource Planning solution, Softengine Coffee One is designed to quickly and easily take your small-to-medium coffee company working at any point along the coffee chain to the next level of success. Learn more about Softengine Coffee One at softengine.com, with special pricing available for SCA Members. Softengine: the most intelligent way to grow your business.

The episode you’re about to hear was recorded live at the 2019 Specialty Coffee Expo in Boston. Don’t miss next year’s lecture series in Portland – find us on social media or sign up for our monthly newsletter to keep up-to-date with all our announcements, including ways to get involved in next year’s Expo and early-bird ticket release!

World Coffee Research is working to improve the genetic potential of coffee cup quality. To start this task, you first need to define the meaning of “quality.” WCR believes it is a market-driven definition and organized protocols to assess quality from the perspective of the coffee industry. With experts and partners, WCR produced the coffee lexicon that allows an objective analytic description of coffee quality attributes. In this presentation, we will discuss these innovations and the lessons learned. Moreover, the latest results deriving from these methods and advances on the molecular basis of cup quality will be presented. In particular, the importance of some specific Volatile Compounds, such as Limonene, will be highlighted. Those results are not an end but a beginning; an opportunity for further studies to find out the molecular markers or genes related to cup quality in general.

Hanna Neuschwander is the communications director for World Coffee Research, a research nonprofit that works to secure the future of coffee. She writes and speaks globally about coffee science and agriculture. Her writing and commentary has appeared in the Art of Eating, Time Magazine, CNN, BCC, Portland Monthly, and many others. She is the author of Left Coast Roast, a guidebook to coffee roasters on the west coast.

Also, I will jump in occasionally to help you follow along.


2:50 Why coffee farmers are in need of high-yielding varieties that also have high cup qualities

Hanna Neuschwander: All right. My name is Hanna Neuschwander, I’m the Director of Communications of World Coffee Research and I’m actually delivering us talk on behalf of my colleague Christophe Montagnon who is our Chief Science Officer. I should just make sure that you all know upfront that I am not a coffee scientist but I am going to do my best to explain the work that we have been doing in the area of genetic improvement of cup quality.

So, what is quality? If what you’re talking about is the attempt to improve quality, First, you have to have some sense of what it is that you’re even talking about, right? What is quality? And of course, quality is determined by you. It’s what the market is ready to pay a premium for. We’re not talking about a standard minimum, we’re talking about a sort of positive definition of quality. What the market is willing to pay a premium for. It might seem obvious to define quality in this way, but this is actually very new in coffee research and in coffee breeding to think about how to improve cup quality through the breeding of new varieties that is looking for a distinct and unique and special positive characteristics, not just a standard minimum quality.

In the past, researchers have tended to regard market evaluation and scientific repeatability as opposites and so they often don’t gets considered together. So what we want to have because we’re scientists, and we’re in the business of breeding new varieties of coffee that we hope will be higher quality. We need to have a way to get scientific repeatability of that market evaluation that says, oh yes, this is something that I’m willing to pay more for. It’s something that I’m willing to pay more or I’m willing to pay more for but that we can have a kind of common agreement about what quality is. So that’s the kind of 30,000 ft. view. It’s going to be important to understand some of the experiments that we’ve been running.

So, do genetics determine quality? One of the cool things about coffee, in my humble opinion, is that the bean is not just a beam. The bean is actually the seed of the plant. It contains the genetic instructions to make an entirely new plant. If you think about all of the coffee that’s flowing around the world right now in roasters and in sacks and on trucks and on shipping containers, every single one of those beans could be a whole new coffee tree, and that does give us some insight, I think, into this question of do genetics determine quality. The answer, of course, is yes. At the same time, the genetic potential of a plant to produce quality, is essential. It’s not sufficient, otherwise, all Geisha, no matter where it was grown, and no matter what you did to it, how you roasted that or how you processed it, would taste amazing and we know that that’s not the case.

So, this is just a kind of illustration of that from a very small experiment that we did. We took 68 coffee samples. There were 29 varieties. They came from seven countries. We didn’t get all the same varieties from every country. It was sort of a random sampling and we did what’s called sensory descriptive analysis. So, this is where we give the coffees to a trained tasting panel and they evaluate what flavors and aromas are actually present in these different coffee samples. It’s a very labor-intensive process and it produced this somewhat crazy looking map.

Heather Ward: Hanna’s is showing a scatter graph with red and blue data points spread across the graph. The blue data points are coffee varieties whereas each red data point is a flavor descriptor. Each red flavor descriptor data point is connected by a line to the middle of the graph, where the X and Y axes meet at zero. Hannah describes this graph as “crazy looking” because red flavor descriptor lines are shooting outwards from the middle in all directions.

Heather Ward: Hannah has highlighted the top right quadrant of the map. It contains three coffee varieties – shown as blue dots – that are far away from the cluster of blue dots around the middle of the graph. These varieties are Pacamara, Catuai, and Geisha. Around these dots are red flavor dots with descriptors of “berry,” “sweet.”

Hanna Neuschwander: Here, let me go in and kind of highlight a few things here. So, it’s not surprising that Pacamara, Where’s my little thing? Here we have Pacamara from Panama and from Honduras. It’s not surprising that Pacamara and Geisha from Panama are doing really well. So, stuff up here in this quadrant are associated with flavors and aromas that we would call positive. Sweet aromas, berry flavor, et cetera. That doesn’t surprise us. It is interesting to see that Catuai from Panama was also very good. We wouldn’t necessarily have expected that.

Heather Ward: Hanna has now highlighted the lower right quadrant of the graph where there blue variety dots such as ‘Costa Rican 95’ along with red flavor descriptor dots of “Dark Chocolate.” The map also highlights that that same variety, Costa Rican 95, is in the lower-left quadrant, with flavor descriptors of “cardboard” and “ashy.”

Hanna Neuschwander: It’s also not surprising that a variety called Costa Rica 95 was in a totally different negative universe due to its genetic differences, and you can see we have lots of samples of Costa Rica 95 from all over from Costa Rica from El Salvador, and they’re all down here with some of the less pleasant characteristics. Are you guys familiar with Costa Rica 95? No, it’s a variety that was released originally in the nineties in Costa Rica. It was resistant to coffee leaf rust, but it was the sort of first in a wave of varieties resistant to coffee leaf rust that was released, and it’s kind of famous in the breeding world because they didn’t do a very good job of testing it for cup quality and it’s one of the reasons why people have this idea that varieties that are resistant to leave don’t taste as good. In fact, they don’t all taste quite as bad as Costa Rica 95 often does, but it is partially responsible for giving those varieties a bad name so not surprising there.

But it’s a little more concerning where it makes you ask a question to see that Geisha from Costa Rica and Geisha from Panama are very far apart and also that Costa Rica 95,  you’re getting a pretty widespread here. I mean, they’re all kind of more on the negative side, but there you’re still getting a pretty wide variation.

So it’s very clear from looking at that very simple trial that we did that genetics are important to the determination of quality, but clearly they’re not determining it universally, right, that the different varieties and different places are doing different things and so every step that you go up, the sort of triangle of determinants of quality is either maximizing or compromising the genetic potential that lies underneath. Does that make sense? Yeah. Okay. So, the genetic potential for quality of a given variety might or might not be expressed in a given environment, so genetic potential is necessary, but not sufficient. In science, we talk about this as being genetics by environment interaction.

How are the genetic potential either expressed or not because of changes in the environment So, we can see that genetics matter for quality but that they’re not the only thing contributing to the creation of quality, obviously. Why should we focus on the genetic improvement of quality and not something else that’s further up that chain?

Heather Ward: Hanna is showing a slide titled “Determinants of Quality.” There’s a graphic of a pyramid, divided horizontally into six slices, with each slice getting smaller as you get to the top. The base of the pyramid is titled “Genetics,” the slice above is titled “Environmental / Farming Practices,” then “Post Harvest,” then “Roasting,” then “Brewing,” and finally, at the top, “Cup.” The vertical arrow on the left points up with some text reading “each step upward may spoil the quality created by the steps below.”

Hanna Neuschwander: So, here’s a pretty simple, straightforward reason. Genetics are the base of the pyramid. If you can maximize the genetic potential, you have the ability all the way through to increase quality that much more but if the base of your pyramid is compromised, you will be fundamentally limited in what you can do further up the chain. From the perspective of the organization that I’m a part of World Coffee Research, it’s also important for us to state that we are a pre-competitive organization. So, we only work on things that happened before pre-harvest or post-harvest processing. So for us, we just aren’t as an organization going to focus on things that impact quality further along the chain.

Another factor here is that the rust crisis and the specialty coffee market. So, the Cup of Excellence many of the amazing initiatives and businesses that you guys have built has really raised a lot of awareness that not all varieties are the same, and that widespread replanting of coffee trees is badly needed in some parts of the world and provoked this other question but with what? What varieties are we going to plant and shouldn’t those varieties, if we can build a higher genetic potential for quality into them? Shouldn’t we? And that really just was not a question that breeders were asking 20 years ago or even 10 years ago. This is a sort of new way of thinking about things.

So, we know that different varieties do have different potential. We’ve sort of talked about that and some of those potentials are higher than others. So, the very classic example would be Geisha but we also talk about SL 28 or Pacamara. These are varieties that have very high potential for quality when they are in the right environment. At the same time, we also know that we do desperately need new varieties. That the old varieties are not cutting it for the 21st century and the problems that farmers are facing today. Most of the higher quality varieties are severely susceptible to major diseases and pests, and that puts a lot of pressure on farmers. If you’re going to lose 30% of your crop to arrest outbreak or a CBD outbreak on your farm, can you really afford to stay with those varieties that give you the potential for higher quality and the hopefully premium that comes with it, but you’re losing 30% of your crop? These are the kinds of questions that farmers are asking and having to deal with all the time, but that we don’t always think about on this side of the value stream.

14:40 Summary of an experiment that identifies desirable flavor markers in coffee plants before they produce cherries

Hanna Neuschwander: So how do you do it? Not a totally straightforward, obvious answer to this, but here’s what we’ve come up with. You need to locate accurate proxies for what is cup quality that a breeder can target because if you just say okay, I want it to be 87. I want this coffee to have a genetic potential to score 87. How does a breeder do that? The answer is, they don’t because 87 doesn’t mean anything. It means that’s how you value that coffee. But it doesn’t give a breeder a target. What a breeder needs is a target that is something that’s in the chemistry of the bean or in the genetics of the bean that they can when they’re making these decisions about how to make crosses and how to make selections among the populations that they’re working with, they can choose the ones that have those particular objective attributes. So, if you can do that, if you can accurately locate proxies that breeders can target, you can save a lot of time and cost to achieve better coffee and I’m going to show you a little bit what I mean by that. So typically, when we’re evaluating cup quality, we’re doing it here.

Heather Ward: Hanna’s slide shows different stages of the seed to cup value stream. She is highlighting the stage after roasting: cup testing.  An arrow goes from the end of the value stream, at cup testing, back to the beginning when the baby plant is in a nursery. As the arrow moves up the value stream towards the baby plant in the nursery, she writes that the time and cost to evaluate cup quality decreases.

Hanna Neuschwander: You already have the sample in front of you. You’ve brewed it, you’re cupping it but if you think about the amount of time and work it took from getting the tree that that coffee came from into the ground all the way through the entire value stream process to bringing it to your table to evaluate the cup, that took a huge amount of time and effort and cost. It took a minimum of three years because that’s how long it takes a coffee tree to mature and just lots and lots of money that’s got to move all over the world. You could also evaluate quality though at the roasted coffee stage. So, still going to take a minimum of three years because that’s how long it takes the seed to grow into a mature tree to produce cherries and it still took a lot of money to get it to your table. If you could do that, you would save some time and some money and the proxy that you might look at would be something like the chemical components in the roasted or ground coffee.

You might be able to look at the chemistry in the roasted and ground coffee, find something in that that is clearly a signal for this is higher quality or lower quality. Okay, that’s cool. You could look at the green beans. Still takes a lot of time, a little bit less money but you could analyze the chemical components in the underlying green coffee to find some sort of proxy for what is quality. You could look at the coffee cherries, still a lot of time, a little bit less money but the gold standard would be can you when the baby plants are in the nursery, when they’re six months old or six weeks old, can you take a little clipping of the leaf, analyze the DNA and say voila, there is a molecular marker inside this leaf that tells me that this plant is likely to produce higher cup quality. That would be the gold standard because you would save so much time and so much money and this would allow breeders to evaluate lots and lots and lots of potential options and home in on the ones that are most likely to succeed.

So, when you’re building a breeding program, you can throw out the stuff that’s not likely to be higher quality, and you can keep the stuff that is and in the process, you save a lot of money and a lot of time. That’s the framework, looking for proxies. Here’s some recent research on that. We recently ran a really interesting experiment. We call it the quality targets experiment where we’re looking to see is it possible to find a chemical proxy for quality? A chemical proxy for quality as defined by expert coffee buyers. You guys are the ones. Coffee buyers are the ones right now that have a lot of power in defining what does quality mean? They’re the ones who are actually putting sort of the money on the line and saying this is what I’m going to buy, and this is what I’m going to sell.

We conducted this experiment in March 2012 and we collected 60 coffee samples that went the whole range of the quality spectrum. Some of them were commercial samples. Some of them were, you know, they were all Arabicas. They were not Robusta but some of them were much lower quality and some were very, very high super-specialty scoring 90 plus points quality. They were representative of this big range of a lot of different geographies and then we had a partner of ours roast them up, and we use this interesting methodology for evaluating quality. We pulled together a panel of tasters, expert coffee buyers from a whole bunch of well-known companies from both the US and the UK. The guy who brewed the roasted the coffee for us and brewed it up. There were three replicates of each of those 60 samples. So, 180 coffees in a blinded randomized order. So, they tasted 180 coffees. They didn’t know which one was which. They tasted each one three times and then all they did was they answered this question for every coffee, and they scored it. There had five options, definitely not specialty, definitely specialty, or somewhere in the middle. We call this the signal detection method.

So basically, we’re asking a single question and we’re getting a strength of the signal in response to that question. Is it definitely specialty? Where is it not so much specialty? Then we ran a statistical analysis of the results and here’s what we found. There were three very clear clusters.  So, all the way that those coffees were rated by the different panelists, and keeping in mind the three replicates, it was very obvious that there was a cluster for the commercial coffees for what we call specialty and top specialty. Really, really clear that there were three clusters, which was interesting, because we had guessed that there would be maybe five. That it would be more clearly differentiated but it wasn’t when we ran the PCA analysis. So, now we have this kind of cool methodology for measuring. We’re finding out if there is consensus around quality, this definition of what is quality using the question, “is it specialty?” So, then we went to analysis here to see if we can figure out what is going on behind that in the actual flavors of those coffees.

So, we took a representative sample from each one of those three clusters, four coffees from each cluster and we sent them to a sensory descriptive panel. These are trained tasters, they’re not coffee cuppers. They’re people that do for a living this for all different food products and what they do is they taste each coffee sample next to a whole bunch of reference food products. So, let’s say they’re trying to figure out is there berry flavor in this coffee? They taste a little bit of the brewed coffee next to it they taste; I’ll use blueberry as the example. Is there blueberry in this? Next to the coffee, they would taste Oregon Fruit Products’ blueberries and light syrup. This is a canned product you’d find on the grocery shelf and a prior group of experts has determined that that product has an intensity of blueberry of 6.5 on a 15 point scale, an intensity of blueberry flavor of 6.0 on a 15 point scale. So, you taste the coffee next to it, and you see, is it more, is it less. Okay, it’s about a 4.5, blueberry, 4.5. So, this is a way of actually giving a score an objective numerical evaluation of what flavors and aromas are present in a coffee. Does that makes sense? Yeah. Okay, and this is using the World Coffee Research sensory lexicon. So, these are the results that we got from that and it was really cool. It worked really well.

Heather Ward: Hanna’s slide contains a table with different boxes colored in red, and different shades of green. If the value of the box is zero, it’s colored red. If it’s colored green, the box has a number that’s more than zero, such as 0.8 or 2. Each column is a different flavor attribute. Half of them are categorized as pleasant and the other half as unpleasant.

Hannah Neuschwander: What we found was that for the commercial coffees, so that’s this group up here, basically, you’re looking for the green. Green is where there is something more than zero for the different flavors and aromas. So, you can see that there’s a lot of unpleasantness in the commercial coffees. This is what you would sort of expect. They might have pleasant attributes, but they definitely have multiple unpleasant attributes. That’s sort of what’s distinguishing a specialty coffee. That middle tier generally have pleasant attributes but often have a woodiness which is really, really common here. Then a top specialty coffee had no unpleasant attributes and at least two different attributes. So, this is really cool because we could see just from the results of the signal detection part that there were three clear clusters but then when we went further and added the flavor evaluation, we can see that there’s a correlation between those results and the signal that we got from the buyers about what is quality. So, that’s pretty cool. So, then we go further to the chemistry. So, those same 12 samples that we sent to the sensory descriptive panel, we sent them to a chemistry lab, and we had them do an analysis of volatile organic compounds. You may have heard coffee is one of the most chemically complex food products that we eat. There’s over 1000 volatile organic compounds that make up the flavor of coffee. So, that’s what we’re talking about. These are the compounds that are generally responsible for the complexity of coffees, flavors, and aromas. We found 50 different VOCs in the green coffee and over 100 in the roasted and ground coffee.

Heather Ward: “VOC” stands for “volatile organic compound.”

Hannah Neuschwander: That’s because new VOCs are created in the process of roasting, so that’s pretty awesome and the cool thing is that we found that there were some VOCs that were, we use the term discriminant, and I’ll explain what that means in a second along these quality clusters. So, here’s a good example. Isovaleric acid. This is in the roasted and ground coffee. We can see that the commercial coffee and the specialty coffee don’t have very much but voila, the top specialty coffee does. Similarly, 3-methyl-3-butenal. There is a clear kind of stepping up as you go through so that’s pretty cool. We also know from the sensory science literature that Isovaleric Acid is associated with kind of cheesy, fruity notes and that Isoprenol is associated with sweet and fruity flavor. So, that kind of makes conceptual sense rate that if you have more of it, it would be higher quality.

So, when I say discriminant, what I mean is that it seems clear that the presence or absence of these is associated with a coffee moving from one group to another.

So, that was cool. That’s the roasted and ground but then this is what’s even cooler. So, when we talk about improving quality for breeding, the closer we can get to the baby plant, the seed, the DNA, the better. So, green coffee is more interesting to us than what’s going on in the roasted coffee chemistry because the green coffee chemistry is closer to the source. So, what we found is that we had kind of similar results for some of these other things but that there was this really, really strong signal for this one particular compound called Limonene, that it was basically not present at all in commercial and specialty clusters but for top specialties is highly present and this flavor in the literature is associated with kind of apple, pineapple. The reason that has its name Limonene is because it’s associated with lemon flavor.

So, that was just one experiment. We were excited about it for a few reasons. It’s the first time we think that we have a methodology that combines an industry expert signal saying, like “Is the specialty or is it not?” with these kinds of sensory, descriptive, and chemistry analyses and it appears that it worked at least to some degree, but we were able to identify a particular chemical proxy for quality, that we can now go out and look in other plants for this or in other coffees for this. So, that’s exactly what we’re doing.  We have a trial, not a trial. We have a, what’s the word for it? A collection of 100 different coffees that are sort of wild type, most of them wild type coffees. We call this a germplasm collection. These are not commercially available coffee varieties. They are the thing that sort of comes before that. The material that breeders use used to work from, and they’re very genetically diverse. So, we have this collection growing in a couple of different environments. They’re just starting to be old enough to start producing cherry, and so we’re going to be able to screen all the coffee from these individual, different genetically diverse plants to see if Limonene shows up in the green coffee samples. So, that’s actually underway. We are doing harvest this year, and our partners at Illy Caffè who do a lot of chemistry research are going to screen them for us, and we’re going to see a) is Limonene there and does it seem to bear out? Does the hypothesis that it’s associated with these higher cup qualities bear out? So, that’s exciting,

The other thing that we’re going to see if we can do, and this gets back to your question, is we are going to see if we can find the location of the Limonene gene. It’s probably not one gene. It’s probably a kind of interplaying series of genes but what we’re going to do is look for molecular markers associated with the presence or absence of Limonene. If we can find that and if it turns out that Limonene really is looking discriminant for quality, then we could do what I talked about earlier. We can take little leaf samples. We can check the DNA and see is this going to be a good bet to keep going with this particular plant in our breeding and we can do this because a group of Italian researchers, including llly Caffè and Lavazza but also a number of Italian universities made public for the first time a full Arabica genome that we can actually utilize for this experiment, and that just happened this year. So, this is research that literally could not have happened six months ago. It can only happen now, and we’re super excited to be working on that. So, that’s one experiment.


31:20 Summary of the current genetics-by-environment interaction research efforts

Hanna Neuschwander: So, I mentioned earlier the genetics by environment interaction. So, we know that genetics alone is not enough to determine quality. We know for sure that the different genetics do interact with their environment and are either expressed or not expressed. So, I’m just going to show you an example of how we know this to be true in coffee. So, this is from a trial series that we have where we have the same varieties planted in multiple environments.

Heather Ward: Hanna is showing a chart that compares different varieties of coffee and different parts of their physical characteristics, such as the plant’s height or the stem diameter.

Hanna Neuschwander: This is from a trial site in Peru, and you can see a whole bunch of different varieties here across the top. There’s about six of them in this data and you can see the thing that this spider graph is showing are the results of measurements around things like height and stem diameter. Basically, vegetative growth, how vigorously is this plant growing. And these results from young plants. They’re less than a year old, so these are not definitive by any means, but they do provide a good signal about whether these plants might be doing well in that environment. So, we can see that there’s kind of a whole bunch sort of in here in the middle that are doing pretty well. Here’s Mundo Novo, a variety from Brazil but that there’s one, wow, that’s like amazing. K7, which is a variety from Africa that, to our knowledge, has never been grown in Peru before, is strongly outperforming the others.

We can go to the same series of varieties growing in a different environment in Zambia, and this is a much different environment. It is very hot and very dry at this trial site in Zambia and we can see that Mundo Novo is still kind of doing kind of similar to how it was doing in the other trial. But that K7 has really shrunk away from that leadership position. It’s fine, but it’s not blowing the other varieties out of the water. This is a very clear indication that there is a genetics by environment interaction going on. If K7 was amazing in both places, there would not be but there clearly is. It has genetic potential to do amazing things, but that genetic potential is not being realized in every environment.

So, the bigger, more interesting question is that’s just vegetative growth, but does genetic by environment interaction impact what a coffee tastes like? And unfortunately, it’s very hard to answer this question because we don’t have any idea what the genetic impact is on quality. We have almost no idea what the environmental impact is on quality and the two of them together and how they interact. We really have no idea. But what’s cool is that we finally do have a way to find out and this is another place where the pieces to do this kind of research were not in place until literally this year. So, we are going to be collaborating with the Specialty Coffee Association and their new Coffee Science Foundation, the UC Davis Coffee Center, and ourselves to do this huge project to try to unlock what is the genetics-by-environment interaction for coffee.

So how does it work? We have this network of trial sites around the world. They are operated by our partners at national coffee institutions in countries around the world and most of them have 30 varieties but we’re actually just going to use the subset of 12 to keep it a little bit manageable and to use varieties that we think have good genetic differentiation in 12 environments. So, 12 different countries. It’ll actually be 16 sites total in those 12 countries, 12 environments. So, it’ll be closer to about 200 samples each year and at those trial sites it’s not like these air just, you know, farmers in their field with a variety garden and they say, oh, this is doing really well. This is doing really well. These are actually scientifically designed in these, what are called Fisher Blocks, so that when we compare the results from one site to the results from another site, we can actually draw relevance or statistical significance from the results because we have controlled as many factors as possible. We also have replicates just like I mentioned for the other one.

Heather Ward: On screen is a diagram of a rectangle sliced into smaller rectangular rows. Each row contains a different variety of coffee. The whole rectangle is then cloned twice, so there are three rectangles in total. This also means the rows are cloned twice, so there are three rows for each variety of coffee.

Hanna Neuschwander: So, each one of these little rows is a variety here and then the same variety might be replicated here and here. So, if something’s going on with a variety in this row, we check is the same thing going on with the variety in the other rows. Is it just a fluke, or is it actually something that’s happening to that variety and we can do that across all these sites.

So, this trial network we started putting the pieces in place in 2015. The plants started shipping in 2016 and they’re just starting to come into maturity and produce fruit for the first time this year. So, there was no platform like this before that we could evaluate the genetic-by-environment interaction for coffee. Now there is and at each one of these 12 sites, what we will be doing is collecting the varieties, processing them the same way according to a really strict protocol, and then shipping them to the UC Davis Coffee Science Center for that same chemistry and sensory descriptive analysis that I described from the other trial.

So, this is the sensory descriptive analysis using the lexicon. We will, we think also be using something like the signal detection method that I described earlier because we don’t just want to know well, what is the flavor in the coffee? Is there blueberry? Is there cherry? We also want to know how do you, how does the industry rate. How do they assign value to it? Is it more valuable or less valuable? So, something like the signal detection method will happen? We think will probably also do the industry-standard evaluation cupping. It’s not fully baked yet, but getting those cup quality scores and then also doing the chemical analysis on both the green coffee and the roasted coffee. So, this is going to produce, as you might imagine, a sort of tremendous amount of data but what it would do is allow us for the first time to examine the relationships between all these different proxies for quality. What’s the strength of the signal that you’re sending about whether this is a high-quality coffee or not? What are the underlying chemical components? What are the actual flavors showing up in the cup? What are the chemical components in the green? And then, if we find enough interesting and relevant here we go and look for what are the molecular markers that are underneath those chemicals. So, we can efficiently target quality in the development of new varieties.

So, we don’t know yet, but we think we might, what is the effect of genetics on cup quality? One of the other things that’s really cool about this trial besides giving us the scientific data that we’re going to feed to the coffee breeders is that all that data also goes back to the countries, and most of the varieties that are in this trial have never been grown in the countries where they’re being grown or if they have, they’ve been grown by a farmer here or a farmer there. But these trials are being managed by the countries and in most coffee producing countries, It’s the country who determines whether or not a variety can be commercialized for farmers.

So, if we are able to find that certain varieties have very high performance, whether it’s for yield or for quality, and we can share that data with the countries and they’re actually generating it because they’re taking the measurements. It means that some countries may suddenly have access to or pursue access to varieties that are much better than the ones that they’ve been growing in the past. It also means that we’re going to be able to see if some of these varieties have environments that are much more conducive to high cup quality. If you think about the example that most people know is Geisha. Geisha coffee and Panama is super famous. That coffee was present in Central America for 40 years. It was growing in farmer fields for 40 years. It just hadn’t found the perfect combination of environment and in this case, also the sort of personality type of the person growing it. It was not a coffee that farmers particularly liked because it doesn’t produce a lot of yield. It grows tall and spindly. The branches break a lot and at the time when most people were growing it, the specialty coffee market didn’t exist.

So, the fact that it maybe tasted different or interesting was irrelevant. It was only in the nineties when Daniel Peterson decided I’m going to try growing this at high altitude, and I’m going to separate the results. The cherries that come from these trees from the rest of the cherries, taste them by themselves, and see if that gives me something interesting and it gave him something so interesting, in fact, that he didn’t know if it was good or bad. He had no idea if people would hate it or would love it. It turned out they loved it. So, this is a kind of nice story of genetics finding their perfect environment and when they did, when Geisha found its perfect environment and the sort of enabling conditions for it to succeed, it created a tremendous amount of value for Panamanian coffee farmers. There would very likely be no coffee left in Panama if Geisha hadn’t arrived or been rediscovered when it did. So, it’s I think a kind of cool example of what can happen when this genetics by environment interaction is understood well, and you can try to optimize the genetics in the right environment. So, we’re going to learn a ton of stuff from this trial that goes beyond just what will get fed back to the breeders for this prospect of genetic improvement of cup quality?

I went ahead of myself. So, it’s going to really, I think, be very, very important the data that comes off this trial not just for breeders and for industry and for countries but also for growers. It’s going to help them answer this question of, if I care about quality, if that’s the market that I’m trying to reach. If I’m trying to be a specialty coffee grower what variety should I grow, given the environment that I’m in, in the particular constraints that I have?  The other thing that’s very exciting about this research is that because we have all these trials in all these different environments, some of them, I mentioned the Zambia location earlier. Some of them today have the kind of climate that we think is going to be more likely in the future.

So, that Zambia trial is very hot and very dry, and it looks a lot like what we think some coffee-growing regions will be facing in 20 to 30 years. So, if we can look at what varieties are really thriving in that environment, those will be good bets for making recommendations for farmers who are replanting because, remember, coffee’s a tree crop. If you put a plant in the ground today, it’s likely still going to be there in 30 years when the climate looks very, very different. So, you want to give farmers good guidance on what is likely to be successful for them in the long term, and right now we have no idea. We are unable to give farmers this kind of guidance because we just don’t have the data.


44:10 Audience Questions

Hanna Neuschwander: So, that is the end of my lecture. We have time for more questions and also for conversation.

That’s a great question. I’m just going to repeat it in case everybody couldn’t hear. So, he was talking about the pyramid. We have this progression of the things that can impact cup quality and his point was that quality isn’t necessarily determined in that order or that it makes it look like genetics has the biggest impact on quality and then environment but that’s not necessarily true. If you roast something in a particular way, you might have a much bigger impact on the flavor than whatever the genetics is. And so, the question is, do we have any interest in looking at which parts of that kind of triangle do have the biggest impact on quality and what can be done about them?

And the answer is yes, we have an interest in knowing that because it’s very relevant to all of these questions, but no, we probably wouldn’t be doing the research on many of those other portions because we are really primarily focused about post-harvest processes. Everything that comes before post-harvest processing. We, as an organization are an agricultural research and development organization. We want to look at what’s happening on the farm and with the plants, in part because it’s a very efficient place to focus your time and energy. Also, because of the fact that we’re pre-competitive and many of the things that… What does that mean, pre-competitive? It means the funding that we get which comes from coffee roasters, coffee importers, coffee exporters. It’s people that would otherwise be competitors are investing in us at a level where the knowledge that we’re generating is not benefiting any one organization or any one company. It benefits everyone equally but once you start getting into research on roasting and research on even processing, it can have competitive advantage. It doesn’t have to, but it often does. So, I think, yes, we’re very interested in that matrix, but it’s probably not where we would invest our effort. But understanding it matters a lot because just to give a concrete example, we have not very much coffee from each one of these trials. We have a kilogram, two kilograms of coffee. If we roast it poorly, we have totally messed up our whole experiment and so, in understanding how to be consistent and how to minimize the impact of these later processes, it does actually have a lot of relevance for the ability to do the research and get the answers to our questions that make sense.

Audience 2: Is there a simple test for limonene?

Hanna Neuschwander: Limonene? Is there a simple test? I don’t actually know the answer to that. I mean, yes, volatile, or I could go backwards. You use the [gas-chromatograph mass-spectrometer]. But it’s not simple. I mean, it’s simple if you have the machine. If it was determined that Limonene was so important, someone might develop that technology. But this is such an early and still not fully baked. It’s beyond the hypothesis, but it’s not yet a conclusion but maybe. But the DNA stuff that’s also… At some point it may be that we don’t actually care about Limonene, what we care about is the molecular marker that comes before Limonene. And then there are rapid-ish DNA assessment tools that you might be able to build in a test for that marker.

Heather Ward: An audience member is asking if Hanna has a complete list of the varieties used in the global genetics experiment.

Hanna Neuschwander: Yeah. I do, not in this deck, but I can pull it up, but I can list off a couple of them. So Pacamara, SL 28, Colombia 3 or 1, I can’t remember. Paranaima, which is a newer Sarchimor-type variety. Marsellesa, which is a newer Sarchimor-type variety. Any more questions?

Heather Ward: An audience member is asking whether World Coffee Research is controlling management for all the international multi-location variety trials.

Hanna Neuschwander: That’s a great question. Yeah, are we controlling management? So, are they being fertilized at the same rate and all of that? To some degree. There are not the same products available in every country necessarily. So, it is a factor and it’s also a separate series of trials that we’re going to be doing that’s built on the similar platform to look at the genetics-by-management interaction and different management regimes and how those impact the expression of the genetics. So, I guess the simple answer is as much as we can yes, knowing that, in coffee, you can never control everything or it’s hard.

Heather Ward: An audience member is asking how many people are involved in this project.

Hanna Neuschwander: So, the international multi-location variety trial, which is what has all these sites around the world. It’s 22 countries have the trials. In each country, there are researchers and technicians that are managing the trial and collecting all the data. So, there’s probably 40 or 50 people involved from the National Coffee Institutes overall, at World Coffee Research we have a coordinator for that trial and our scientific director as well. So, there’s that and then the sensory analysis and that component, that trial hasn’t started yet. So, we’re just getting the coffee now, and we’re building the scientific plan for that and we think it will probably be another 10 people or something. So, we’ll have a graduate student at UC Davis and a Post Doc at UC Davis, probably a few other people on the chemistry and sensory side running those components of the trials and then doing data analysis as well. But we’re actually with the Coffee Science Foundation. They are actually in the process of fundraising for that right now.

Heather Ward: An audience member is asking who are the partners in Brazil for this trial.

Hanna Neuschwander: For that trial? It’s not in Brazil. We have some Brazilian varieties in the trial. So, for that particular trial because of the nature of the trial, we always go through the National Coffee Institute. We don’t work with private farms for that trial and there are politics involved with what we call material transfer, moving genetic material around to different countries. So, we’ve stopped and started those conversations more than once. We’re very hopeful that it will happen at some point, but there’s just been more politics involved there than in other places. There are a few places that we don’t have this trial, that we would love to have this trial, but it’s a process, and I said we started in 2015. We actually started in 2012. It took three years just to build the relationships and the trust and the infrastructure just to do this. Running a trial like this, in this many countries with moving plant material around is incredibly complex, incredibly political and I think really one of the biggest successes that we have at World Coffee Research is the fact that this even exists. It’s very unprecedented that this many countries would work together to share their plant material and hopes the plant material developed by other countries. So, if that’s the only thing that ever happened that we did, I would still be really proud. Great question.

Hanna Neuschwander: Thank you for being here on a Sunday morning.

52:20 Outro

Heather Ward: That was Hanna Neuschwander at the Specialty Coffee Expo in April 2019. Remember to check our show notes for a full episode transcript of this lecture and a link to coffeeexpo.org for more information about this year’s event.

This has been an episode of the SCA Podcast’s Expo Lecture Series, brought to you by the members of the Specialty Coffee Association, and supported by SAP’s Softengine Coffee One. Thanks for listening!

Subscribe to the #SCAPodcast on Anchor, Apple Podcasts, Google Podcasts, Spotify, Breaker, Castbox, Overcast, Pocket Casts, PodBean, RadioPublic, or Stitcher.