Building Trust in Science with AminoChain

Transcript

GMGM. Welcome to what's going to be another fantastic episode of Devantel. So, if you didn't know, Devontel is a 30-minute podcast held every week allowing founders, hackers, and anyone in between the opportunity to come on the show and showcase what they built. And today, I'm ecstatic to welcome Casper Barnes, who is the founder and CEO of Amino Chain. So, if you didn't know, Aminochain is a community-owned infrastructure built to make research faster, more transparent, and more inclusive. So, if you stick around for today's episode, you'll get to meet Casper, learn all about amino chain and the origin story around it. All right, let's do this.

[Music] >> GMGM, welcome to the show, Casper. I'm excited to have you on today, man. >> GM, no, thank you for having me. I'm I'm very excited to be here. >> My pleasure. My pleasure. Uh yeah uh I know you are a busy man uh and uh are joining us from a conference live so really appreciate you taking the time out of your day to do that. Um just I guess uh for the audience today u if they aren't familiar with yourself would you just like to give a brief introduction?

>> Yeah absolutely. Um so thank you for having me now I'm super excited to to speak with you today share a little bit more about who we are and what we're building. Um we are currently at ASHG American Society of Human Genetics. So big conference with all sorts of different people in the bio banking space and you know clinical genomics and research um human subjects research ethics and so on community engagement all the topics that are relevant to what we're building at immun chain is like a good you know few hundred if not thousand people here and so on. So that's the exciting thing we have in front of us. Um but yeah so as a quick overview of you know who I am and what Amino chain is. Um my name is Casper.

I'm u the founder of the startup chain originally from Cape Town, South Africa. That's why I've got this funny accent. And um I did my undergrad degree in the United Kingdom. Um studied economics and neuroscience and moved over to the US around four years ago. Um I did a graduate degree in biotechnology and worked in cancer research for a while and then did another graduate degree in bioeththics at Harvard Medical School and founded the Mina chain around three years ago. And initially when we set out to to to build a mun chain, we were focusing on building a NFT marketplace of donated stem cells and we were addressing the issue of not having enough allergenic stem cell donations

for people with blood cancer. And so for people that don't know, if you have leukemia, you have lymphoma, any type of hematological malignancy, a blood type of uh disorder, one way in which you can get cured is through something called allergenic stem cell transplantation. And that's where you go and you give a crazy amount of chemo and radiation to a patient has blood cancer and you kill all of their blood forming stem cells and then you take healthy stem cells from a donor and you put them into the patient and it engrafts into their bone marrow and then it reboots their immune system. And we need roughly 15,000 transplants every year for patients. And we're only achieving around 12,000. And that's through public registries of

donors. That's through using umbilical cord stem cells. That's using hlo identical donors. So patients that are from um or donors that come from a sibling or parent, for example. Anyway, around 3,000 people don't find a match every year. And those patients unfortunately pass away. And the more we looked into it, it turns out that there's a huge economy for donated stem cells. So when people donate these stem cells, the nonprofits, the registries are actually selling the donated stem cells for $40,000 per allergenic graph. And that means these nonprofits are turning over half a billion in yearly revenue selling other people's donated body parts. And that blew my mind. I say that makes no sense. How is there this

huge economy for donated material that somebody gives out of the kindness of their heart? the more we dug into it, it turns out that consent rates are particularly lower amongst marginalized communities and patients of color. And so, um, on the registry, if you're a white patient with blood cancer, you might have a 75% chance of finding a match. And if you're African-American, you have around a 28% chance of finding a match. And if you're Hispanic or Pacific Islander, you're just shy of uh 50% likelihood of finding a match. So, it kind of opened up like this really interesting use case of there's a donated there's an economy for donated body material, people giving this stuff for free and people selling it and the people that are hit the hardest are the ones that don't have

that trust in the system, typically marginalized communities or underrepresented populations. And so, we decided to build a system to democratize this with crypto. We said, "Well, how about we just tokenize donated stem cell graphs, turn them into NFTs, list them on a decentralized marketplace, and then let, you know, researchers come and buy these NFTs, and when they buy them, we pay royalties back to the patients that donated them, and they can track where they go." And that was the first thing that we built. We thought, you know, this distributed NFT marketplace is a good way to, you know, bring back agency to research participants. And, um, it was a phenomenal idea. I still love it.

We we submitted a prototype in a hackathon to get going. I met some great engineers. We put together our founding team and that has now through a crazy process of iteration, lots of speaking with bio banks and you know community health organizations and so on. That's evolved into a general you know blockchain infrastructure to build healthcare applications. We we are first focusing on connecting bio repositories of all types of donated body parts. Um and we're not working with NFT standard anymore. We have a different scalable architecture, but um yeah, effectively what started off as the idea of a marketplace for donated stem cells is now turning into a vision of building the Ethereum of healthcare.

Wow, that's amazing, man. And yeah, that's that's quite an undertaking. And I mean uh for all the the the that crypto gets for scams and rugpools and all these things, seeing the tech the seeing blockchain blockchain technology used for actual um potentially life-changing uh purposes is is is phenomenal to see. Um, and yeah, I I I read in a um a blog article of yours that you guys are actually partnering with Syndicate um for your blockchain in infrastructure. Um, did you want to just get into that a little bit and and the purpose behind that?

>> Yeah, exactly. So as you can imagine with um something like you know what we're building at the general purpose blockchain for healthcare apps there's a lot of you know real world hurdles that you have to overcome just like a basic example would be um working with bio repositories or hospitals if you want to sell them blockchain based software describing the the the the tech stack and the systems and getting the hippo accreditations and um explaining effectively how the whole software works. into a provider agreement or into a collaboration agreement or anything that a legal um office of a hospital can review is incredibly complex. So the first time we tried this with lawyers,

they just said, "Yep, everything's going to be visible on a public blockchain." And you, you know, take full responsibility of understanding how tokconomics work and what validated nodes are and so on. And obviously the second we gave that to a hospital, they were like, "What the hell is this? There's no way we're signing those." Um, and they were super confused. So we had to take a lot of time and effort and diligence into designing a really intricate protocol that you know bridges onchain you know verifiability and trust but also allows for scalable offchain computation and flexibility in where PII and PHI is stored so that you actually achieve your regulatory standards and after a lot of research and you know

speaking with the phenomenal research team at ACC crypto our investors and others we've now landed on a sophisticated design that bridges an appchain infrastructure on syndicate where we can control information that's publicly visible. We can control you know gas and token prices. We can um basically sequence a transaction in a decentralized way and still leverage the trust of crypto but have a lot more um flexibility and you know and security and building trust that we need with the the healthcare institutions. So, we're excited to partner with Syndicate on that.

>> Wow, that's Yeah, that's amazing to hear. And yeah, there's there's always um this weird friction with having um this type of I imagine sensitive data held on chain and a lot of people um publicly available on chain and a lot of people who aren't accustomed to that or accustomed to data living in a centralized location kind of have this oh what the heck what do you mean? Um but I I mean I I think data like this in particular I think it makes sense to to house on chain and you guys seem to have uh uh cross those hurdles uh at least at the beginning um to kind of get to somewhere that makes sense and I guess uh going beyond um stem cells that um

you mentioned as the initial version of amino chain uh what other kinds of um scientific data are you guys planning on storing? Yeah, absolutely. So, um you started off with um stem cells that you mentioned, but um it's like it's important to note that there's a very distinct economy for different types of biomeaterial. So, like for the world of what we just called their allergenic cell donations, which basically means stuff that comes out of a healthy person and into a sick person straight away. For example, blood blood donations you donate and then transfusions happen straight away. Stem cells, same thing. You donate them and then they get transfused into a um patient. Again, that is a very, you

know, specific value chain. Um you have a way to recruit people, you have things that are stored in a way, you have uh standards and making sure there's no infectious diseases and then you you get reimbursed medically by you know including that as an item in the bill of the patient when they have the transfusion. So that whole universe is totally separate from um the world of IVF for example, sperm and egg donations, right? That that is its own universe entirely, but it's similar in that body parts are stored and they're used for a procedure and then they have an outcome. So the world of um organ donation and the world of stem cells and the world of IVF and the world of

disease tissue, the world population genomics, they are all totally separate. um they have very very different you know flows and payers and users and um long story short what we are focusing in right now is the world of what you call diseased tissue. So that means people that have um some kind of disease perhaps cancer or something else and they donate a biopsy sample as part of a routine part of surgery for research purposes not for clinical purposes. And of those types of bioreositories that store diseased tissue, we have roughly 2,500 in America. Um, and across those 2,500, there's roughly 200 million samples

stored. And across those 200 million available, we're only using around 10%. And then another 90% of samples that are stored ready for research use are just sort of lying dormant and people can't find them. Um, and even if you do find them, there's long processes of trying to acquire them. And so the first thing that we set out to build with Mina Chain was basically a search engine for these existing bio samples. And that gave us a foot in the door and a relationship with all these big provider networks. And um, we've grown that out now to around 30, you know, bio bank strong uh, network um, across the United States and um, other areas of the world like Europe and Africa and so on. And um across these 30 different bio banks, we have roughly

410,000 bio samples and we have data on around 150,000 unique patients. So scientists can come and they can search across this distributed database of stored diseased tissue that they want to use for research and they can request that directly and then it can get shipped and then we increase the visibility of existing bio samples. Now that's got nothing to do with crypto just yet. It's just a totally centralized solution, but we've got a great foot in the door with all these big hospitals now. And the next thing that we want to build is the same users that are interacting on the search engine where they go and request the samples, we now want to flip the switch and actually make them interact with digital assets on a chain or with

something on a digital record, right? Something that has a blockchain back end. And in doing that, we then, you know, unlock a whole world of new user experiences for these scientists. they could, you know, return information or they could, you know, query um unknown, you know, clinical data on a deidentified research participant or they could build a clinical recruitment registry or they could do all sorts of things in a decentralized way because the infrastructure provides in the back end. So, um yeah, long long answer to your question, but basically we're focusing on on um all sorts of disease tissue.

Yeah, that's that's uh that's wonderful to hear that so much data is now available for scientists and researchers to to use. I guess um um having this much data available um is there any um scientific breakthroughs or or anything like that um that have come from this data being available that you can share or maybe in the works? >> Not not yet unfortunately. Um so the the platform was launched maybe like five six months ago now. Um but the the network has grown quite strong. We have um recently signed with some very large academic medical centers um in the United States. Some um I think I can share them. Yeah. So we

signed with you know providers at Northwestern for example and um some at scripts and some at um um McGill University and um you know a number of others and basically now all these scientists can go and log on and look through the inventories and uh distribute their samples but what exactly they're using them for we don't we're not we're not tracking just yet. So that would be another step as well. >> Gotcha. Gotcha. Gotcha. But I mean the ball is rolling. So um you know scientists are going to get to work right when they have when they have that data available to them. Uh and I guess this is uh a good kind of segue um if you wanted to kind of share the the

platform itself. Just kind of take us uh on a mini tour of uh what amino chain looks like the platform. >> Yeah. No, I'd be happy to. Um if we could pull this up. So, this is You can see my screen. >> Yeah, you're good. Sorry, I was on mute. Yeah, you're good. >> Amazing. Um, great. So, this is the Spasman Center. This is totally live. Anybody that wants to can go and log on and check all the stuff out. Um, and if you go to app.aminochain.io, this is what you see. A whole network of bio repositories that we've integrated.

And you can click on the profile page of the bio bank. You can see the high level overview of collections. They have their contact information. So totally transparent and this is effectively like Facebook for bio banks. People can go create their profile pages. You can see the overview of their collections and some of them we've integrated into their inventory management system. So we're pulling live updates about how they use their samples, what they have stored, what's available. and others they send us a um inventory list on a regular cadence like once a week and then we take all that information and we update it here on the specimen center.

And now this is what the view looks like for filtering across all those collections. So of those 30 different banks that we have on there here you can see um the filters that we're able to search by. You can search by diagnostics, by specimen type, by donor information or so on. Or the most recent thing that we've just launched is natural language processing. So you could try search for something as a normal string. You could say, let's see, we want breast cancer samples from women under 30 in the USA.

And then this beautiful thing called AI will ingest the query. and we'll attach the relevant filters um that apply to whatever we're looking for. And then it'll return all of the results of the available samples across the whole network of 30 bio banks. And here we go. You can see there's around 125 specimens that fit our need. And we can filter through all of them. Here we can see pathology information, preservation details, and this is all deidentified, right? So these samples aren't necessarily linked back to the patient. But if we wanted to, we could add them to an inquiry and

then we could request the specimen directly from the bio bank. So we could put information here saying, you know, did this patient have a one mutation or whatever you want. And this is live. So, if I send an inquiry, then that would actually go to the bank. So, I'm not going to do that. But, um, this is what the flow looks like at the moment. And this has got nothing to do with blockchain. What we're excited to do now in the next few months is actually launch this as a protocol integrated product, which means that this 29year-old woman, even though we don't know who she is, she could have a patient interface and she could track

how her samples and data are being used. And when I buy this specimen here on the marketplace, I can pay royalties back to her. So she can earn dividends from the use of the specimen. And because there's a connection back to her, we could query any survey information she fills out or any clinical data that she has on herself without knowing who she who she is. So we could query that through clinical record uh through zero knowledge groups um or you know other encrypted systems that we have in the back end. And that is effectively where the blockchain comes in. Um, and this is where we are right now.

>> Amazing. Amazing guy. It looks like you you've built quite a bit and uh it's just a few months there. Uh, I'll bring your screen off. Uh, yeah. I I guess um is there uh since we are a uh developer oriented show, um are there any plans to um expose this data via an API or or something like that in the future? Any any use cases you guys have thought around that? Um, yes, absolutely. Like I mean the people want retrospective samples, right? Like they want what we what I just showed there, but people don't buy samples to look at them because they're beautiful and they want to hang them on their fridge. They want to, you know, turn them into data sets, right? And so

that's where the real value is. Um, and that's what we're here discussing at this conference, right? Like I mean people want not just genomic information, but they want proteomics and cabalomics and transcrytoics and so on. We need all layers of what's happening in your molecular biology um and in your geneticology. And so that is what people really want. So that's what we're moving into now. Like instead of just looking through existing biopecimen collections, we will actually start to aggregate multiomic data sets on the same people and then still have this economy where they can like earn incentives and they're actually sharing in the benefits that their donations derive. And um you know when all of that data is aggregated into the system theoretically yes if it's a decentralized protocol anybody should be

able to source that data and build a target discovery platform or use it in a way that's you know decentralized and we would just be the permissionless infrastructure layer to let people do that. That is of course our ambition. Like practically what does that look like in a step-wise action is a question that I guess we're still answering because once you have the data sourced, do you have one dedicated trusted research environment? Do you let people download the data and then it's gone and you lose the provenence of it or do you let anybody integrate their own uh trusted research environment at the front like surface the data. Um there's still a number of things that we need to consider in there but the you know the vision and the ambition of course is to be as decentralized as possible which mean it would be akin to something like

an API people you know sort the data or do their own pipelines or discoveries or whatever on the stuff that's in the decentralized bio bank. >> Yeah. Yeah. And definitely yeah it sounds like there is quite quite a bit um to kind of still shuffle through but um it looks like you've built or you guys have built quite quite the foundation already. So, it's going to be exciting to see how it kind of evolves um evolves from here. And um is is any of this open source by chance? If if there are some people here who are interested in in helping continue to build out this vision uh for them to contribute or I guess are you guys hiring if anybody is interested in uh in uh throwing the hat in here?

>> Yeah, absolutely. No, I mean we're certainly hiring. Um absolutely hiring. Um we are you know deep in protocol development um design something that will be progressively more and more decentralized um and eventually we'll get to a point where yeah it'll be open source people can build on it at least that's our ambition right um and you know in the meantime unfortunately the software on the back end isn't open source but the interface is and the system so anyone can go and um use the data or the application and Um we are always on the lookout for top talent that wants to come help in and you know join our mission in building trust in

science and healthcare. So please do reach out if you think that's you >> and I I guess on that point if people do want to reach out uh what's the best way for them to get uh in contact with you? >> Absolutely. Our website is um the best you know www.aminoain.io. Um, we are also active on X and LinkedIn and newly Instagram. We just redid our Instagram page and need some love. So, anybody that's active on Instagram, please go give us a follow there. Again, just amino chain. Um, and yeah, so through the website, through our socials, please just reach out. Like that'll go directly to to us and our team.

>> Amazing. Amazing. Uh, and I got a couple more questions here before uh we let you go. Um just wanted to ask you um in terms of the roadmap for Amino Chain um is there anything that you kind of like to share or kind of tease out um for what's upcoming perhaps um early 2026 or or uh maybe even this quarter? >> Yeah, absolutely. So the the next thing that's on our agenda here, right, is like all the things that I just showed you don't actually have a connection back to the human being that donated them, right? And now we're actually starting to focus on the stuff that we came here to do, which is, you know, bridging equity and agency back to the research participants. And so we're so

freaking excited to roll this out. But people will basically we in partnership with two different perhaps even three different um sites and bio banks that we're speaking with at the moment. We will have kitting infrastructure set up and um we will send these little kits to people's homes. There'll be a small capillary blood collection device in it. People can collect blood, put that in the box and send that back to the bio bank. Oh, sorry. When the kit arrives, they have um an app on their phone, the donor portal that they download, and they scan the kit ID with their app.

There's a cryptographic handshake happens there between the kit and the donor portal. We send the kit back to the lab. They bank the blood. They generate data sets, populate it in this beautiful bio bank, and then and then we're live. Anybody that comes and buys that stuff, the patients earn points and they and the, you know, this pointing system they can then exchange for a suite of different incentives, you know, for Uber credits or for flight miles or for whatever else they want to exchange it for. And um we're testing out the whole incest in incentive system, the whole kidding architecture, the aggregating samples and omix. And the two main indications that we're focusing on now will be psoriasis and myastia

gravis. And um that's that's on the on the road map now. We're recruiting for those actively. So I should have actually mentioned that as well. If you know somebody that's um suffering from one of those two conditions and hopefully we can we can we can be of service to you and and um get you more agency in the research process. >> Beautiful. Yeah. There you go. uh if you're someone watching today uh and that kind of fits your uh is a fit for you uh definitely reach out to Casper. And uh just to end things off uh kind of on a general uh general note uh I just was curious to kind of get your thoughts on um D the DIY space in general and where you where you think it's at and

where you think it's going to be heading in say like 5 years time, 10 years time. >> Yeah. In other words, when token, right, is >> not quite not quite just in terms of like um we know there's like a lack of hardware and whatnot for a lot of people that want to run uh big model sets through uh ML processes and whatnot. And yeah, I guess how do you see how do you see kind of this lack of resources or innovation in the space kind of um evolving over time?

>> Yeah, it's a really great question. Sorry, I was being a bit facicious there. Just like jokey, but like a lot of the DI narrative today is like inextricably linked with whatever crypto asset people are trying to launch like against the you know projects. And so I think um if you're an early developer and any crypto enabled product, if you launch a token a token before you have product market fit, then you seriously risk losing credibility in what you're building. Um, and so you need to design what you're doing carefully and get users and already set up as much as you can, even using stable coins if you have to, but then move into a bit more of a, you know, token guided network effect or whatever. So, um, what does that mean in the healthcare? Because that's like a 10

time harder problem, right? Especially if you're trying to service institutional healthcare or enterprise healthcare customers, it's it's, um, it's a messy situation to be in. So um I our ambition with Amino chain is to be that composable layer that has the relationships with all these big um academic medical providers and big final companies so that others can go and build DAPs on it that there's automated compliance and that there's you know trusted infrastructure to everyone that if let's say for example they wanted to build a deep pin network for contributing healthcare data towards ML like you just said or if people wanted to um I don't know build um a um a system for tracking your medical records with

you between providers that you go to whatever like you would need the different providers to all be linked to the same blockchain infrastructure. So anyway, we want to provide a platform for developers to be able to build that, right? That that's our goal. And um I think DI in general right now is suffering from a little bit too much of a of a like crypto first narrative as opposed to a innovation first narrative. Um, aside from like the hardware things that you mentioned, I think people in general are are too like entrenched in trying to do everything in a hardcore decentralized way from the ground up.

And not many builders have like try to approach this with like a prudent mindset about progressive decentralism and you like how do you compromise on needing to be a central authority for trust a little bit in the short term so that you at least gain trust and then you can make a permissionless trust and so on. It's like there there's there's a serious art to building a company in general, but then trying to, you know, build a crypto company for healthcare institutions is like really freaking difficult. So, um, you know, where do I see the DA ecosystem in general? I I I have the same vision as everybody else.

I I totally believe that we'll get there at some point. You ask me in a hundred years time, would all medical records be centralized? Would there be no way to track my data from one place to the next? would I have zero incentives in you know all the research data being used for research but like no obviously those those problems will get fixed they have to get fixed right so in a 100 years I think we will address them um I think there'll be many attempts of people trying to fix it and not being successful in the meantime because they approach it in the wrong way and I think that's ultimately what we're trying to be very careful um and you know very deliberate and hopefully if we do it right then it will be a win for everybody because we could just hand

over the keys to developers and let them build to their hearts content. >> Great answer and a great uh uh great thing to kind of leave our viewers with um to stew on a little bit uh over the weekend. Um Casper, uh thank you so much uh for taking uh the time out of your busy schedule to come chat with us today. It was a pleasure having you on and uh yeah, really looking forward to seeing uh how Amino Chain um evolves um uh as as the months and years kind of go by here.

>> Thank you so much, N. had a lot of fun. Really enjoyed it. >> Excellent. So did I. So did I. All right. Well, uh with that, just want to wish everybody a very happy Friday, happy weekend wherever you may be. And we'll catch you back here for another great episode of DevT next week. Till then, have a good one, folks. Bye.