#136 Joe Meyerowitz

Scientific training really rewards sort of breadth and novelty, the ability to show many possible things that you can do. And a company needs at least initially one product that is decisively useful to a specific customer or customer type. You know, there needs be clear buyer painful problem something you can produce that solves the problems and advantage over alternatives favorable economics.

often what I see in science entrepreneurship is people produce these amazing systems and they say, look at all of the different things it can do. But there's no one single thing that does so much better than what would win in the market.". Thank you for listening to Sprint – The podcast from the Federal Agency for Breakthrough Innovation of Germany.

I'm Thomas Rammgen and very much look forward our guest today Joe Myrovitz who leads a new challenge at Sprint called Engineering Microbial Consortia. Joe was a biologist and engineer who's worked at the intersections of science, for example, at Caltech in Francis Gray Institute. And entrepreneurship also known as building companies... In which he was very successful.

Now it's Brent! He and his team are looking for future science entrepreneurs that can fundamentally change how we produce chemicals and materials with biology. Joe thanks for your time. Yeah great to be here. Let's start with the basics. What is a microbial consortium? So microbial Consortia are communities of microbes, you know small bacteria and yeast in things like this.

And when they grow together When they live and interact in the same environment They show some helpful properties. So nature is very good at adapting to changing conditions, if the temperature outside changes and these microbes are in the soil or If things that are available eat change These groups of microbes can respond in ways single microbes cannot And so we're very interested in seeing if there's a way to engineer these kinds of microbial consortia the ways that they are naturally robust to changes in their environments and use that for manufacturing.

Give us some context, why does this matter? And why aren't single micro fermentation processes good enough? Well so much of the chemical and materials economies still depends on fossil carbon And current biomanufacturing, the use of biology to produce chemicals and materials has really only been successful for a limited set of products.

In a lot of cases they require being fed something very consistent in refined like a very refined sugar. These sugars end up being too expensive to use. that we would want to produce from sustainable sources instead of from petrochemistry. And so if you just look at Europe's existing pool of refined sugars and starches, it might be about ten times too small for the demand in materials in Europe alone.

And so we need to find ways to access sort of a much wider range of sustainable carbon, and if we use microbial consortia there's the chance that will be able feed these processes. A lot different kinds. changing sugars in changing materials for them to eat lower costs than way it would radical also by enduring changes in environment might be used with lower cost infrastructure.

And so this has to do with sort of making it feasible for the first time, to produce some these chemicals that are currently made from fossil carbon today and start making them with sustainable inputs. Oh is that mainly a matter of price now? Or would these new biological materials also replace stuff we can't... produce biologically correct right now?

So I think there are two opportunities with microbial consortia. One is to reach new costs, and that's by using lower cost infrastructure and using lower-cost inputs in a sustainable way. And then there also the possibility of using the division that you get in these sort of microbial ecosystems to reach more complex products, that maybe we would want to produce with biology but haven't succeeded with our current approaches.

And so I think there's a hope for producing new high value sorts of chemicals and materials—of the sorts that…that we really want uh...to produce sustainably from the get-go. replace the existing chemicals and materials that we don't make sustainably today, find a new way of manufacturing those. Give us some examples?

For microbial consortium specifically using them for manufacturing. there are number of lab examples but they're relatively few examples kind of manufacturing I have in mind which is in market to day. so one useful example might be adipic acid for a chemical product. This is common commodity-chemical with an established market, and there are number of different inputs that could be converted into adipate acid.

in the research realm you can see we make increasing amounts from this valuable molecule but nobody has really demonstrated the ability to produce it in commercial quantities. And so one of the hopes of this challenge is that we can find teams, have breakthrough innovation and how to use microbial consortia To produce a product like this commodity chemical adipic acid.

Another example Is something called NAKER. This is like mother of pearl, the sort of shiny material inside Moisk shells. And there was a demonstration again in the lab using different microbes to lay down the different layers of this complex composite material and they were able to produce very interesting potentially high value synthetic materials with these microbes But they used them one at a time.

And so, they would use one microbe to put down one layer and then different microbe for another layer. If someone could find the way potentially get all of these microbes together To build this structure in sort-of-one go it might be possible to produce enough that will be commercially useful. I think there are different opportunities in chemicals or materials here where a consortia would have a clear advantage in what they'd be able to do.

that hasn't been done yet. What applications could benefit most? The biggest opportunities are, again... ...to find lower costs for existing chemicals we already use when producing them in the sustainable way and also opportunities to find more complex materials can't make commercially with biology. And so I think there are opportunities in textile industries and materials coatings, potentially optics a lot of different domains where currently sustainable manufacturing is minority-of the market.

if we get that to majority then eventually one hundred percent it will reduce our Dependence on imported carbon and the release of CO² for our materials world. How can AI & machine learning now leverage what you intend to do? Obviously, AI & Machine Learning will be a game changer in this context as well!

So how exactly should people working on the subject use AI to invent these consortias? Yeah, so I think the real opportunity with microbial consortia is to create robust manufacturing processes. Those that can use changing inputs and still give us a consistent output. And microbial Consortia demonstrate this in nature already.

but the challenge here are very complex and they have a lot of moving parts. Different members of the consortium may be performing different chemical tasks at different times, uh...the relative proportions of each member of the community maybe changing over time. And so I think that real opportunity with AI and machine learning is the ability to predict and control these.

So even within all those changes inside the community and the changing inputs in a changing environment, you can still get consistent manufacturing output from this system. And so one way to think about it is maybe there may be just few underlying variables for these systems that capture behavior that matters for manufacturing.

So we could think of an attractor. There are thousands of mechanical parts in this big system, but in the end at our level there's a shifter and it says forward neutral or reverse. And then actually only a few choices that need to be made to direct its activity into specific direction. So I think if we use AI machine learning these were tools what are called low dimensional latent spaces, or these simplified views of this very complex systems that would allow us to see.

What all the sorts different settings for microbial community and then keep them in ones we need from manufacturing? Interesting I assume it already happens. how successful has AI already been applied in the field. So I think there are good examples for AI being used to demonstrate two big tasks, and these are sort of grounding believing that we can use this consortium for manufacturing.

One is with a natural community like one who naturally lives on waste water treatment plant. You can use AI and machine learning, some experimental data to show that these natural consortia can be steered to produce one particular molecule or another particular molecule. This steering happens at a very high level.

it has to do with how much they're being stirred in the temperature and by connecting those external variables with the output that you're looking for, there are ways of seeing how you could get a good manufacturing process to come out of this system. And one or other way is in composing new consortia from scratch where you take individual microbes at different types.

I think that these microbes could grow together and if they grew together, they would be able to do certain things. And there have been a number of examples in just the past few years. for example screening thousands of yeasts showing specific pairs are able to grow together by exchanging some key metabolites where maybe one can make something needed And the other one can make something that is needed for growth, but they can't make those themselves.

So by cross-feeding with each other They're able to Stably grow. and so I think in these synthetic consortia There's a lot more control over this system But it's also much more complex. These are both domains where if we build on the existing results will have some real opportunities in manufacturing.

Interesting, so who should apply? Well spring really backs teams that are prepared to build. So we have a call for applications planned to come out early October and The deadline is planned about six weeks after the call opens sometime mid-November and it'll be a short online application followed by an in-person jury pitch for selected applicants.

I would say if you have scientific or technical insight that could become a significant breakthrough, uh real ambition to build around it You should consider applying. It can be as early as Uh applying and saying We will form a company, it can be at very early stage. And I think what really matters the most is clear core team A team that's sort of prepared to lead and credible view of commercial opportunity That you think we'll support.

taking this technology all the way into real world It doesn't have necessarily been a technology you've invented. I know that there have been examples in Sprint's recent challenges, where sort of non-traditional participants who've come and been able to do really exciting work. There was a movie director who did particularly well on the past software challenge related to deep fake detection.

And so I think if you have a unique perspective on the problem and you can identify the right pieces of the technology, then your vision for what we want to build. Um...I would still encourage you apply! Read The Call. it will explain what were looking forward and hopefully uh..it'll be an opportunity really move some these technologies into real world use.

What's the funding mechanism? Sprint Through its challenges can fund staged R&D without taking equity and so this is generally staged in three parts. So it's a competitive challenge using pre-commercial procurement And you submit cost estimate, and then there's a fixed price that's agreed around that cost estimate.

Right now, we are expecting to be able to fund up to one point two million euros in the first stage which will run about fourteen months. One point three million euros uh... In Stage Two Which Will Run About Twelve Months And One Point Eight Million Euros. In Stage Three Which Will Also Run About twelve months.

and so there's Up To Four Point Three Million Euros Per Team Across All Three Stages plus. that were applicable. But there are caveats. The current plan is for up to ten teams to begin Stage One, but then up to five teams continue to Stage Two and three teams continue Stage Three. And the selection of which teams will have to do with progress you've made towards milestones that we asked.

I think something that maybe is useful to think of in this context, is this very much for building something new. And so... ...I would hope the teams with think-of themselves as time constrained not just resource constrained. and So if you think about it a month has may be about twenty working days and so two day delay In your project.

use up about ten percent of your productive month. And if you are using about a hundred thousand euros per month, which this challenge would support starting in stage two... A two-day delay could cost around ten thousand euros. and This is the rule of thumb not careful accounting calculation but The ideas that the teams that move fast and Are able to reach sort of significant evidence that they're on the right track are the ones who most likely to continue into future stages and get funding in later stages.

Joe, I hear the entrepreneur in you speaking actually! Being optimistic as we are would be a very good outcome for this challenge. So what's our outcome? are companies coming out of this challenge, even if they don't get to the final stage ideally hitting the right points of evidence and the right milestones.

And sort of resolving the right obstacles that there. ready for the next commercial step whether that's early sales product submitting engineering samples to customers so customers can test what the ability to raise additional capital where it's needed. The hope is what comes out of this are sort of sustainable companies that can rely on the market, keep moving these technologies into real world use and I think alongside what the outcome could be is by looking at the end-to-end process that's proposed.

And so, what materials do you think are going to start from? What will they do with a consortia and how can they use AI productively in order for them to make sure these consortias produce a valuable manufacturing output? Who needs it? In which form or quantities does one need it? And all of these things need to be considered together.

So there will be an emphasis, as we're watching the applications come in and then helping teams that are selected... ...to ensure that the techno economics or process makes sense.. ..and meaningful life cycle analysis so we understand how well this end-to-end process can work in context at a broader market.

I understand that we expect a lot from the people participating at The Challenge, which is normal. challenges should be challenging. What are the hardest problems to crack? From technological point of view first?". So again wearing my entrepreneur hat for a minute here... Possibly the hardest problem technological.

There's a lot of evidence at the research level that these sorts things are possible, and getting them to work on large scale commercially viable is going be challenge for any biomanufacturing technology. but I think possibly the hardest problem will choose right product understanding what they can produce and because there are certain products for which you might need to be producing thousands of tons before any customer will consider buying them.

And then, they'll be other products that potentially... You could reach much faster with the simpler process as a first product one or maybe several where you may only have to produce a few kilograms. And so I think this combination of understanding the technical abilities of the process that you're building, and then understanding what the market wants together is actually going to be the hard part.

Interesting interesting Joe give us a little bit background on yourself? We can feel how deep your technically in the field, how big your knowledge on entrepreneurship within or at the crossroads of biology and company building is. What drew you into the field? And what did finally made sort going this step towards entrepreneurship?

Sure! So I started as a physicist an electrical engineer. I thought it was really interesting that there are parts of the world That, Are very difficult to describe and predict And work with Just from first principles alone. So if you just sit down With the equations Then write everything out It doesn't actually tell You what's going To happen.

Chemistry & biology i think Had This very Attractive example Of this Where they're Very essential For big Parts Our modern Civilization. And yet a lot of things, you just have to go try them. To figure out how they're gonna work. and so Richard Feynman has this concept. what I cannot create? I do not understand...and that's very useful way for approaching these sorts engineering problems which is perhaps we don't truly understand microbial consortia.

but if creating new consortia and creating new manufacturing processes with these consortia, then maybe we will be able to sort of understand them better. And I think this is again one of the examples where if you just told me i'm going to mix these microbes together tell me what's gonna happen? But maybe we will be able to in a few years.

And I think being able do that sort of, In the context where it's useful for humanity is very valuable. What would you have liked? To have known on science entrepreneurship before becoming a science entrepreneur? scientific training really rewards sort of breadth and novelty The ability to show many possible things.

company needs at least initially one product that is decisively useful to a specific customer or customer type. There needs to be a clear buyer, painful problem something you can produce that solves the problems an advantage over alternatives favorable economics. and often what I see in science entrepreneurship is people produce these amazing systems and they say, look at all of the different things it can do.

But there's no one single thing that does so much better than what would win in the market.". And a lot of cases... The first company does are most important ones. All those other things you could do aren't really an important part but the very first thing to choose from them Interesting and sounds very logical.

Why do a lot of science entrepreneurs don't have the hunch to do that? Well, I think... A lot of scientists learn how to do this along with the process. but you asked what would i like to know before we started? it took me time actually building things then trying to sell them and get them into the hands of people who really needed this thing that I was building.

Before, I saw it only need to be good at one thing... At least in the beginning. And if i'm as good with a single thing It doesn't matter all these other things. maybe I could have done With tools like those hard-earned knowledge, and so I'm saying it out loud in the hopes that maybe they'll save somebody else a few months.

No no i didn't want to question you. obviously make sense but...I mean..but you know what? It sounds totally logical once you've understood but probably something that is hard to know when you start off. I mean, scientists often are people who are used to thinking complex systems and so forth. Obviously it's no wonder many of them will not focus from the start right?

Sure! And i think we all find a particular joy in thinking three four five steps down the road uh...and building a company sometimes only practically can take one or two. And though you should have some long-term strategy, sort of the things that you can address and sort of actively get better at are the ones right in front.

Makes total sense. last question to Joe which I pose all my guests until now what radical innovation or what Frank's renovation do you wish for? In the year twenty fifty? You're not allowed say something with bio manufacturing. Oh, that's hard. Okay so I'll say it and you will tell me if i break the rule about not having to do with biomanufacturing but its my personal belief.

sustainable energy is becoming more abundant. they renewable sources supplied over fifty percent of Germany's gross electrical consumption in twenty-twenty five uh... It being adopted at an increasing rate. And so, I think we're looking at a future where energy intensive and sustainable can be compatible again.

This is not something that we've had in the past – We have had energy-intensive industries and we have had sustainable industries…. …and if we have enough sort of primary grid electricity... ...improved grids and improved storage all these different things together We might be able to start thinking about sustainability in a way where we're not asking how do we use less To-do more but we can just use more and Do more?

In a way that's not harmful. And so I think there is some real examples around maybe direct air capture of CO two, and Direct conversion of co two into different useful molecules and Different ways doing manufacturing, which might include the things that we're talking about today to really produce a future of material abundance in addition to I think the future that we are headed for.

Which is a future sustainable energy abundance? I can accept it as an answer and i'm very grateful all the brilliant answers you gave us on this interview. Thank You Joe! Thanks also to you, dear listeners for listening to our podcast. If you like it please leave us the rating in your podcast app and we might also recommend this or other episodes too.

friends of yours The Sprint Podcast is produced by Max Zonner. Thank You Max. as always at next episode will be out in two weeks And until then Please As Always Stay Curious.

Kapitelklicken springt direkt hin

Shownotes

Transkriptvom Podcast bereitgestellt · Zeitmarken springen hin