C. Schafmeister | Hardware and Software of a Critical Path to Bottom-up Molecular Nanotechnology

okay everyone, welcome to that group of

molecular

machines, really, very happy to start a new year with you and the group is now and yeah, I think it's really the third iteration, so it's really exciting, and I think Yes , you're definitely gaining momentum, it's growing, the applications keep coming in, so thank you all so much for nominating such wonderful participants and for nominating such great speakers, we're really great for the first question and I'm so excited. for our next workshop as well, so maybe just to get people talking who may or may not know exactly what we have waiting for you, we have in terms of our seminars, so

molecular

non-cat eleven can always be found on our site Web.

We have Krishofmeister today so it's already on the schedule and we've had a lot of really wonderful seminar presentations in the past, feel free to dive into them and they all have a seminar summary, but if you're interested in what we have in store. for you and the rest of question 1, you can click here, if you're interested, if you're looking, it's a new group and you want to join, this is the apply button and then the second thing I want to tell you about. It's our America Design and Systems Workshop coming up, we're already starting to plan it.

We had a lot of wonderful time bonding with each other, so if you want to meet Chris in person, you should be inspired by today's talk from him, so that's it. really great way to do it, please start applying and we are definitely accepting applications on a rolling basis and already, um, those are the top few and two announcements, if we have time, maybe I'll introduce you to the interns after the talk, but for now I'm really excited to have a question that has definitely been a friend of forecasting for a long time and, you know, I think usually whenever we write our previous seminar reports and it always went straight to your section. which a lot of people skipped um and uh that's been really exciting so thank you so much for coming and you're from Temple University.

He'll be discussing candidate hotline

software

on the

critical

path

to

bottom

-up molecular

nanotechnology

and hopefully maybe some snippets as well. about all that there so I'll be in the chat I'll be monitoring the questions um and then I'll give them to you later so I feel free to keep coming like crystals starting to prepare and I'll devour them. them up and then um later okay, okay, wonderful, well, the stage see you thank you very much for coming and I'm very excited for the talk, okay, thanks for the introduction, can everyone listen to me?, okay, can you hear me, great, okay, let's try it.

I think the pedal, do it, okay, that full screen now is cool, okay, just turn this down, okay, so I'll tell you. I'm going to look forward to a lot of things we're working on and some proposals. I have things I want to work on and this is a rough summary of the talk for someone to tell you how we are going to deal with future pandemics using our technology, how we are going to cure aging, how we are going to provide abundant green energy and how We are going to promote molecular

nanotechnology

and I know it sounds like a lot.

The first one is already supported. It has been supported since 2019 by the Department of Defense. Reducing threats to defense. agency this is the organization that protects the United States protects us from chemical, biological, radiological and nuclear threats. They are funding us to the tune of approximately 12 million dollars and I thought I would start with what we are doing there to sort out. to uh, show you what the basics of how our chemistry works, um, okay, so, um, I actually have two groups of people that are physically working about 45 minutes apart, uh, this is my lab at the Temple University, um, these are my students, one of My postdocs, this photo was taken a couple of days ago, up here on the top left, they have developed much of the basic chemistry that has made possible the technology that we're developing on third law and then this is my group on third law molecular uh two of these people are from Temple University um they're all chemists and a programmer and uh they're also developing this technology uh that's funded by the Department of Defense we have a board of directors um basically how This started when I went to the Department of Defense in 2019 and I said I think we could develop a technology that could replace monoclonal antibodies and diagnostic tests and that we could rapidly develop these things if a new threat emerges, a new virus that you know, appear from time to time.

We could quickly develop molecules that put them together and then turn them into diagnostic tests, so they said, "Cool," but we don't get anything from academic research, so we want you to start a company. What we've done is more of a research organization currently funded by the Department of Defense, so here are the basics of how the upper ligamers work. We've developed these building blocks, they're rings and they have two groups of amino acids and uh, over about 10 years we figured out how to synthesize these things and how to put them together into larger structures. We had a few dozen of these molecules determine their structures, but they didn't do much because they didn't have any functionality.

I have no side chains apart from the inherent structure of the core molecule. I'm just going to do a quick sound check. Can people hear me? Yeah, yeah, okay, so in 2010, we figured out how to introduce functional groups into the building. blocks so that each building block has a side chain and then we figure out how to connect them with these functional groups to create these ladder molecules that we call spiroligomers. Well, you're probably all familiar with DNA with proteins, peptides, um, these are different. The essential difference is that these building blocks are joined together through pairs of links and therefore you get a rigid and complex three-dimensional structure that can be decorated with all these different functional groups and presented in 3D space. in different directions and then, um.

What we've been doing since 2019 is putting these molecules together to create really big molecules. They're about 5,000 Daltons, they're about the size of a very small protein, and in the last three years, we've created two and a half million of these molecules and tested them against dangerous organisms that were provided to us by the Department of Defense and identified molecules that bind to those organizations. Just a quick summary. I'm not going to go over all of this, just a quick summary comparing spiral ligamers. to DNA like DNA origami versus proteins you already know with protein design, the people who design proteins these days.

Sphero ligamers have some additional advantages. On the one hand, they are much more structured, they are ladder molecules and we can join them and then cross them to create. uh, little packages that are where you can basically control where each atom is in three-dimensional space. They are extremely stable. They do not unfold like proteins in DNA. Can we put thousands of different functional groups on them compared to DNA which has four? chemical bases and proteins that have 20 amino acids in the natural set of amino acids we can put thousands of functional groups into our building blocks.

Spiral ligamers are now manufactured with robotic organic synthesis and are quite inexpensive to manufacture. $25 a gram for the monomers and we could reduce that cost to virtually zero with some technology that I'll show you in a moment. Now these molecules have many applications and we are really focused on developing applications, so the application that I am The first thing I will tell you is this simbuzz and the idea is to join three of them to form a molecule that is like a claw that can wrap around a protein and bind it and bury as much surface area as an antibody can.

We have created two and a half million of these molecules and have shown that we can quickly identify the ones that bind to proteins. Another application is to put them together in packages with pockets and put metals in the

bottom

of those pockets and create artificial enzymes. catalysts, they could do things like sit up, create synthesized solar fuels, cleave cross-links that build up in our bodies as we age, and potentially address the diseases of aging. We can make pouched molecules that could be active sensors for all kinds of specific small molecules. like neurotoxins, we can make therapeutic molecules, we've created some protein-binding molecules and I think this technology will serve as a kind of way to push nanotechnology and molecular technology forward, so I'll show you some of these how it works.

So the first project is currently funded by the Department of Defense. We've been there for about three years now. We just passed another milestone last night and we've received about nine million dollars to date working on this. The idea is to create these really big claw-shaped molecules that can replace monoclonal clonal antibodies and then incorporate them into lateral flow assays like the Covid tests we've all been doing for the last few years and develop tests for different types of different organisms. that they are indestructible that they do not have to be thrown away every six months that they can be stored for 10 years, taken off a shelf and they will still work and, most importantly, that they can be developed in just a couple of weeks given the infectious agent and begin testing in months and millions of tests in about six months, that's the mission here, so this is kind of the basic idea of ​​how it works.

We have here a white protein called trypsin, the red one is the fingerprint. of an antibody that binds to trypsin, that's how much contact area you need to make to have one protein behind another protein, well, this symbol here, this molecule with three segments that I've covered just to show an idea of the scale and we are making very large molecules that can bury as much or more surface area than an antibody and that is essential to bind proteins selectively and tightly, so the first thing we did to make this possible was to increase our synthesis, we outsourced the synthesis. from a late intermediate stage of our building blocks, the four compounds, one, two, three and four, for wushi aptech and synthesized 12 kilograms of these four precursors in separate barrels, so it took them six months.

We have four barrels of these intermediates. I've used about a third so far and, yes, it's perfectly pure. What we then do is process it further and incorporate all these different chemical functional groups. These are all small pieces of known medications. They are small. The polar heterocycles that are known to bind to proteins and that are known to be safe as drugs are part of the drugs, so we show that we have about 22 of them right now and we make two forms of these, the block of central construction here has two. stereocenters and we make both the RR and the SS shape of these building blocks, these are mirror images of each other, so we have 22 functional groups in two different backbones for a total of 44 different building blocks, this is right where we start , there are thousands of clusters that we can place on these building blocks here is an image of some of these samples, on the left are the RRs for the building blocks, on the right are their mirror images and we have tens of grams of these, so what we do is uh We choose random sequences like 60 sequences for the week that we load these robots with, you know, a couple of grams of each of the building blocks and then every day we make 12 of these molecules here at the bottom and the way this works is just like solid phase peptide synthesis, it's like robotic peptide or DNA synthesis, you add a building block to a resin and you remove a protecting group, you add another one, you take away one protected group, you add another one and you attach a molecule with the sequence you want, then we hit it with an acid that separates it from these plastic beads and removes all these protecting groups on these groups and then promotes a reaction in which each building block attacks the previous one and forms a second set of amide bonds, this works wonderfully. 700 of these molecules at the last stop in about eight months we manufacture about 60 per week, when we do a production run, we manufacture 10 to 100 milligrams of each of them and each of them has a specific rod shape. with different functional groups sticking out in different directions in space and I have also developed

software

that allows us to build three-dimensional models of these molecules and now we are working on obtaining crystal structures of some of them, so what we did was take thesemolecules three at a time and we put them together in many different combinations, we created what's called an encoded DNA library and we created two different forms, these two different forms, each of which contains about 885,000 members, where we mixed them together. and we pair different segments in different combinations, so these are these claw-shaped molecules and then we take these libraries of the compounds, each one of them, they are in these little rhythms the size of a human cell 10 microns wide, there are many copies of one of these we call them symbols synthetic molecular binding agents uh there is a version in each bead and there is also a DNA tag that encodes the sequence then we mix it with a fluorescent protein uh we mix a million beads with a fluorescent protein and the beads that have a molecule that binds to that protein become fluorescent and light up like a Christmas tree light and the ones that don't bind to the target are black, then we run them through a device called an activated cell sorter. by fluorescence that separates the fluorescent beads from the black ones and you can shoot, currently we process about a million and a half beads per day and we isolate the beads that have bound to the protein and then we go and PCR amplifies the DNA, we perform a sequencing of high throughput of the impacted beads using alumina sequencing and then we identified the hits that emerged multiple times over several days, the actual hits and then we resynthesized them and validated that they bind to the target.

So here are a bunch of beads showing Simba himself and they all light up because they've attached the fluorescent target organism. This is a bit preliminary, but we've done this three times now and gotten results each time. um and then we take some of these punches, we remove them from the beads, we spray them in nitrocellulose and then we load them or I send bacteria on them and the dry molecule islands capture those bacteria, so now we're developing, we're trying. We are working hard now to develop them as diagnostic tests that do not involve monoclonal antibodies.

Well, I'm going to change the subject here. I want to talk about something else that we can do with this technology, probably the most important thing and that is. catalysis, so something that's not really appreciated is how nature builds things. Nature builds things by assembling these wonderful magical things called catalysts called enzymes, organizing multiple groups in space and joining together metals in many cases and creating pockets in which other molecules can react and get products. and the catalyst does not change with this, so one catalyst can produce billions of product molecules, this is how nature builds us, it builds everything in nature, um, we have done this, we have created individual spiral ligamers that accelerate chemical reactions like this one we designed here. to do what's called a transesterification reaction, it converts this pesky vinyl trifluoromethyl into a methyl ester and it does multiple changes and as you know it speeds up the reaction about 2500 times, this is nothing to write home about, it's not a very impressive catalyst. .

Enzymes, biological enzymes can speed up reactions billions of times and this is something that we need to develop as a species, this is a capability that we really need to develop is to rationalize the design of the catalyst because if we could, we could do things like create molecules that split water to produce electrons, powered by solar energy to produce solar fuels to raise carbon dioxide to alkanes and produce solar fuels from sunlight carbon dioxide and water this is something all green plants do we could create if we could create stable catalysts that did this we could create completely green fuel solar fuels, um, that would solve all of our energy problems, we could make catalysts that would help us make our building blocks for Spirit ligamer synthesis, so that a barrel could be thrown into a little proline and, a couple of days later, take out bisamino acids and release them. the cost of synthesizing the basic components is essentially zero.

Catalysts could be created that could selectively modify existing drugs and create new compounds that are not commercially viable at this time. All of this could be done if we could create molecules that had pockets with metals attached to them. inside them and the pockets control which substrates can reach the metal and react with it. Here's an important application for this as we age, proteins in our extracellular matrix react with glucose and sugar and form these cross-links called glucose paint. This builds up through aging and we don't have natural enzymes to remove it, it's probably responsible for many diseases of aging, it's definitely responsible for wrinkles, the hardening of tissues, you know, the pain you feel in the morning when you wake up at night. as you get older.

I feel very compelled to work on this project. The problem is that these cross lengths are formed in very small spaces within the extracellular network and even if we found biological enzymes such as bacterial enzymes that could cleave this group. they won't be able to get into tight spaces because they'll be too big, they would also generate a violent immune response if you inject them into people and there are a lot of reasons why this isn't going to work very Well, so I'm not going to get into this, this is just kind of information on advanced glycation and glucose pain products and how it's most likely related to aging.

What I would propose is that we take Spiegel's total glucosamine synthesis. which has been published by Dave Spiegel's lab at Yale and we make a molecule like this that has a fluorescent donor on one end and a quencher on the other end. This is a common thing, it's called a molecular beacon and then we developed small ones based on spiromer. catalysts that can enter the extracellular matrix and selectively cleave them without cleaving anything else. What it would look like is four, three, or four spiral ligamers linked together in a package with a metal tied inside and a pocket that would match and selectively cleave glucosapine.

The way we would discover this is that we would make hundreds of Sphero ligamers with different metal binding groups, assemble a DNA-encoded library like we do now for Simba and treat them with the glucose pain molecular beacon and then look for the ones that generate a fluorescent product, they select those beads, identify the catalyst and then study how they work and begin to make a computational design to improve them. Switching gears again, another thing we can do with spirit binders is put them together into little bundles. which are like little bricks about two by two by two nanometers with a lot of control over the functional groups sticking out of them and we can use cross-linking groups that would allow them to self-assemble or, through positional assembly, assemble them into more complex structures up to complex machines, so these are just cartoons, but I've been working for the last 20 years to develop software to design these things and I've gotten this to work now, it's kind of like Rosetta, but not natural. building blocks where you don't have a protein databank to build and what you can do is about a million lines of code, it has its own built-in compiler and allows you to specify building blocks using this type of syntax, here you specify what your folder looks like this, this completely describes the Spirit ligamers, our folder for describing amino acids and proteins, it only needs three lines, it only needs three entries, but the higher ligamers are much more complicated and then what the software does is take. these descriptions and then run uh and I've been doing this for the last few weeks um it takes about 1600 CPU hours to build a set of 1200 training molecules and these training molecules are just a building block in a context of a pair of other building blocks and then they are capped to make a chemically reasonable molecule and then it does a lot of molecular mechanics conformational searches throwing in things that you know where you get the bad convergence, pulling out all the unique shapes that these training molecules can take and then accumulating all these unique shapes in a sort of trained database and that allows you to build large structures very quickly which we can then look for structures that might contain groups in particular three-dimensional contexts, so this will help us design. proteins spiritual ligamers that bind to proteins that create active sites and catalytic sites and ultimately create machine parts in 2007, long before we could even put functional groups on these molecules.

I wrote part of the roadmap for nanotechnology with the Foresight Institute and the Battelle Corporation and the um. um oh, the guy who did Gateway Computers um The Family Foundation I'll remember soon um there they were, we had this roadmap for molecular nanotechnology. I wrote a chapter and described how we could do a positional assembly with spear ligamer bundles, this is what I did. You would place these positions and essentially mimic the Stuart platforms with synthetic chemistry and then use positional synthesis to assemble them into complex structures that could ultimately become machines that could build even more complex structures.

So, that's all I have right now. I wanted to finish, we have, we are developing, we want to develop all these different applications. We are currently working on a similar project, but there is a platform technology that can solve many problems, so I will stop there and be happy to answer questions. Thank you very much, yes, we are all here and no, no, we are all here. I think you've definitely stumped some people here and we already have a ton of questions in the chat um and this was wonderful and assuming we can delve into a member in some of these slides.

I love that you took a little risk on some things that you could delve into, that was wonderful, thanks for that um. I think maybe we'll start with Logan, uh, with you first if you want to mute and ask a question and, um, yeah, from now on if you want to put your question in the chat or raise your hand, they're both fine, both, right ? Um, so I was wondering about the low cost of spiritual binders and specifically in comparison to DNA. I was wondering if that figure I think the DNA was used at a thousand dollars a gram or something and then the spirit binders were 25 a gram.

I was wondering if the DNA figure was drawn from DNA that is produced by chemical synthesis or if it also includes DNA that is produced by biological turnover. Okay, so you could, yeah, yeah, then the cost comparison would potentially be more similar. if you included DNA that you would first synthesize chemically and then amplify using PCR or grow in bacteria or something like that, no, what I'm talking about is chemical synthesis of DNA and that's just a number that I've heard if anyone wants. To update myself with a better number, I just made it this morning and that was hard for me, so, yeah, look, the great benefit of DNA and proteins is that we have amazing, wonderful catalytic biological machinery that can put them together, but you know . uh you know, so I don't have that with the Spirit ligamers, sure, yeah, and I can definitely see the DMX benefits of the Spirit ligamers.

I was just curious how they compared in that sense with that kind of additional qualifier um uh. there, but yeah, thank you, that's very informative. I appreciate that thank you wonderful we have a Korean name. I don't hear anything foreign, yes I'll stop, yes please go ahead, so yes, two questions, one is, what role? Does simulation play a role in your work, particularly in active site design? If you do, what role does it play? What types do you use? And then the second question is: hey, I noticed that the language looked a lot like a lisp dialect, how did you do it? happened uh so answering them backwards are common lists that I'm using as our kind of programming can do is a common lisp that interoperates with C plus and it's about half C plus half common list and we write the common list compiler from scratch. um, how much do we use simulation?

Not much yet. I've been working for 10 or 20 years to develop the software to get to the point where we can use it for modeling. So far we mainly use plastic models, but I plan to do so. Use it a lot. I wrote jump, which is an interface to Amber that about a quarter of the world's biomolecular simulations go through every day to set up. It's a program that sets up calculations for biomolecular simulations, so I use it a lot. To use molecular mechanics, we are properly using the amber molecular force field, so it's not like that, I mean it's okay to guess acommit, but not so good for designing a live site where all the electronic details have to be correct, that's what I can do is great, very cool, I wrote a common compiler from scratch, huh, that sounds like a long time and effort, yes, it's open source, it's been on the internet for years, it looks like it can do it for developers on GitHub, it has a huge fan base. and I think exciting Creon is definitely a big feat, so congratulations Chris, next we have Stefan.

I just want to say that I spent years programming on Commonless, so I'm fun and excited. Well great. Look, I'm really looking for more. people to work with us and if you're interested contact me yeah we have a small team that's been working on how it's actually going to open up because we can design these molecules with this amazing software. Next time we'll have Stefan Bosley, yes, thanks, that's a great talk. I'm sorry. I mean, I'm particularly interested in the machine side. I mean, can you incorporate, you know, build these systems so that they can display recorded amplitude controlled movement. you know, for example, a kind of hinge type behavior or rotation of one component in relation to another because obviously that's important if you're thinking about building machines like Nano, yes, flexibility is easy, flexibility is something we have to work on hard to avoid, so building molecules that have hinges is simply putting a single bond between two rigid units and you have a hinge.

We published a molecular actuator that joins metals and opens and closes them and we demonstrated it 12 years ago, like 10 years ago, it is very easy to create things with high amplitude movements. It's difficult to create molecules that are highly structured and that you can connect together to get those high-amplitude controlled movements. Yeah, yeah, cool stuff and then my other question was with these types of polymers. You have all these different types of drives that you can put in, that seems like a really interesting potential storage polymer to me. Have you thought about it from that point of view and do you know, for example, if you tried to read the sequence of these structures using nanopores so that you can sing or something? which you can't because they are complex three-dimensional structures and a lot of the information is buried in the stereochemistry, which is something that is um uh you can't extract information from them um and sequence them because you can't reduce them to a linear structure once you've assembled them. , yeah, sorry, I'm just thinking before you put it together, obviously, but yeah, that's fair, thank you very much, yeah, even before no, let's start with a sequence that we want us to do and show that it has.

With the right mass, we're working on growing crystal structures so we can see a three-dimensional shape, but you can build the three-dimensional shape of the plastic model and it's complex and rich because of all the stereocenters you can. be one of two different configurations, the symbols we're making have 30 stereocenters and if I make one of those, I think it's almost impossible to figure out what the stereocenters are after the fact, I mean, it's actually kind of a thing. we could use it for intellectual property protection protecting intellectual property um uh the only way to really get the structure out is by growing a crystal and then determining its three-dimensional shape okay, okay, thank you.

I also just shared the Google-only Nanotech Broadband is here in the chat that you referenced and then we have Ted Kayla. I can't hear anything, Ted, can you unmute me? Otherwise, I can read here, yeah, yeah, so I'm excited because, uh, The key thing, to make an analogy with the semiconductor industry, is to get on the

path

to profits, so that once you can make a product that generate profits, banks are willing to lend and things can take off, so I'm excited that You have two different potential products here, both the disease diagnostic and the catalyst, so it's really exciting.

I'm very glad you're doing it, yes, and talking about someone with a company with a burn rate of three million dollars a year. You know, developing this technology is something I think about every day. um, we have, I mean, I think there are huge commercial applications, I don't think there are huge commercial applications for this because now we can produce these molecules very easily and design them for applications. um, we have to go out, convince investors, convince people to support these kinds of things, um, and that's what we're working on right now. I'm working on a business plan, working on which we're also driven by the dod um uh milestones and yeah, it's been a White Knuckle journey the last three years to get this technology off the ground.

This is, yeah, I'm also impressed with the stability of the things that he's created for disease diagnosis, um, you. I know for pathogen detection, the fact that they can stay on the shelf for a long time is really good, right, thank you, yeah, that's important, that's something that was very important to the Department of Defense, yeah, well. thanks, well I'm going to dig a little deeper, maybe into similar long-term applications. A new path there in a second. I just want to make sure we get to one more question in the chat, which is from me, the driver, and her. you have two questions, number one is how do you get it from the Lego instant camera that, for example, the enemy makes, and the second question is how do you focus on link types instead of DFT-type exploratory simulation.

Thanks, the comments and how to read them would be a little like this. For me, thanks for reading that, okay, so how is that different from Lego's chemistry as an enemy? So, enamine comes together, it has a lot of different building blocks that come together through single bonds and there's rotational flexibility around all of those single bonds and then you create a molecule that's a little bit flexible, it takes on a lot of different shapes and you know the principle Fundamental to structural biology is that structure defines function, if you have a molecule that has a function, it has to have three: dimensional shape to achieve that function, so flexible molecules can bind things together, but they have to freeze much of it. of his movement to do so.

Now the chemistry that enamine uses to bond things is the same chemistry that we used to want whether we used to bond. segments together and that we link we will link modules together with um uh basically we want to use an efficient high performance chemistry that works all the time a reliable chemistry, so we use the same type of chemistry that enamine uh uses um, it's just the difference is we have rigid building blocks that are held together through pairs of bonds and enamine has more flexible building blocks that are held together through single rotatable bonds.

So how do you focus on the link type instead of the DFT type? Exploratory simulation, so we use Amber I. I'm one of the Amber developers, you know, I wrote Jump and I've been using it for 25 years, 30 years, um, it's the only way we're going to be able to create a lot of different shapes and quickly look for commits that we use. quantum chemistry to parameterize our models, but our models are kind of a classical model, so using DFT and Quantum is something that we use to parameterize them and we use quantum mechanics to model transition states of reactions when we want to create a catalyst that can stabilize a state transition, so that's a place where we combine quantum mechanics with molecular mechanics, so I'm agnostic, it's just that quantum mechanics is very, very slow, it scales very poorly, like to the third power, the number of electrons what do you have. uh, so you can't really deal with very large structures. um I'm really excited about these new deep learning methods, um, that can predict quantum energies from structures.

I'm watching it very closely, it's something I would like to incorporate. uh, so I'm agnostic, whatever works, but it takes pretty sophisticated software to build molecules properly, uh, parameterize them correctly, run millions of simulations, organize them all, get results, that's what I can do, it's wonderful, yes, I am particularly you. Obviously, I'm biased because I also run biotech in the longevity group, but I'm particularly excited about potential longevity and aging applications and I'm wondering if you could talk a little bit more about that and you know, diving. maybe a little more about what's possible here, there are open challenges, which might get people excited.

I know, for example, you don't have to speculate too much, but you know, like doing this, maybe in the forest, uh, yeah. It's very relevant to the longevity community because I think a lot of the tools that the US machine group people are building are really relevant to the tools that the people in our biotech longevity group are looking for, so it would be very anxious. I want to hear you explain a little more and I think the most important thing in an extremely important application to develop is to be able to rationally design catalysts to produce molecules that could produce other molecules because then you can create the building blocks that you need to do everything.

Nature's starting materials, which are carbon dioxide, nitrogen, oxygen, water and sunlight, this is how nature builds everything that we need to develop that ability to produce artificial catalysts, this application of glucosapine, cleave glucosepine It's something that interests me especially because I'm getting older. feeling the pain every morning the clock is ticking and we don't know exactly what the cause of aging is. I mean, I saw a new one a couple of days ago which is, you know, methylation, there's some evidence that methylation patterns in DNA but fundamentally, I think any cause of aging is going to come down to chemical bonds that form or They break in the wrong place and the way to fix that is with catalysts that could get in there and change those chemical bonds back to what you need them to be in storage.

The glucosepine example is just one of those. It definitely accumulates with aging. It definitely makes your tissues tighter when we were 80 90 years old, about three quarters of the length of the cross in your extracellular matrix is ​​Glucosapine we have no way to cleave it, so I think one idea is to make 10 million potential catalysts that have different pockets and then throw them into a molecular beacon with glucosapine. You will find molecules that contain glucosapine. You could throw in a lot of other valuable or similar starting materials to those that build up catalysts and pull out a lot of other additional catalysts that do other reactions and then figure out how they work, figure out what their shape is, start modeling, design better ones. and learn how catalysis works and then begin to rationally design them for specific reactions.

Amazing, wonderful, very cool, and could you maybe talk to some of the other parts of the slides, as well as draw a little path, maybe for Therapeutics? Well, yeah, so Therapeutics, so we've made it like the individual spiral ligamers, the little segments with four, three or four functional groups, ten stereocenters, um, with just those that are the molecule size, they're about a kilo and Dalton medium in size, we have designed some. of those that bind to a protein called mdm2 and we published that in 2012 and the compound enters the cells, it binds to the protein and in fact stabilizes it, causes its levels to accumulate about 30 times inside the cells, those molecules could be medicines, it could represent, I think so. represents a new and rapid pool of therapeutic compounds, we know how to make millions of them and then quickly identify the ones that bind to the target protein, find out what it looks like, how it binds to the protein and then start developing better ones.

Medicinal chemistry, um The compounds themselves have, I think, good pharmacological properties, they won't be chewed up by proteases, they won't generate an immune response. We can create thousands of variants of them and quickly sift through them to find better, uh, ones we've been looking for. to enter cells by passive diffusion. I think this represents a huge new reservoir of therapeutic compounds. We need to start exploring this. Okay and maybe a few more words along the way because it has a lot of different applications. and like almost most of the videos, so we have groups, so I guess a lot of those students might be interested in that, so I didn't understand the first word, a few more words about sensors, like sensors, top right . a little bit on your slide, if you could, explain a little more, like you know where you see the potential applications first.

Do you already have any ideas? Yeah, things like neurotoxins, like shellfish toxins or teratotoxins, small molecules that are produced by shellfish or dinoflagellates, um, or things like polyfluorinated firefighting chemicals,you know, forever chemicals, pphases, uh, fentanyl, opioids, lots and lots of molecules, we would like to be able to selectively detect them at very low concentrations, we can create molecules that are like triangular and that target groups. the inside and create pockets and imagine it's like a little triangular box where you can change the shape of the inside if you make a shape that is complementary to the molecule that you want to attach like fentanyl, then it will bind to that thing if you attach those triangular molecules to an electrode surface that you know as a surface acoustic wave sensor or a chemical field effect transistor, you can turn that kind of traceless sensing platform into a selective sensor for that chemical agent that people are trying to do this. antibodies, but antibodies develop when you put them on surfaces, they don't last very long, they keep their binding site several nanometers away from the surface of the electrode, while the molecules we would produce would be just below the surface of the sensor, here there's a huge application for making long-lasting environmental sensors, you know, chemical sensors for all kinds of small molecules, amazing, cool, before I go and literally go down the rabbit hole of each of these individual offerings, we have to Micah, uh, who stood out quite well with another one. question, so you may have already answered this, but please answer again because I made a mess.

Go ahead, proteins have the protein folding problem where we don't know how they actually fold, unless we use complicated AI. similar problem where we don't know what form it's going to take until we build it or have the ability to easily verify it. No, you can take, even someone who hasn't had good organic results, a university organic chemist can take a Plastic Molecular Modeling Site that you can get in a bookstore and tell me what each of my molecules looks like, any spiral ligamer, um. , in a couple of minutes, based on a molecular model, um, they are easy to predict their structures. but they are still complex and rich because the stereochemistry rotates the groups in different directions, so it is programmable.

So no, we don't have a folding problem with individual segments. We know what they look like if you hinge them together like we do. Now that we've started doing, they have some moves that they can do, but they're pretty restricted and with what we can do, we'll be able to search those spaces just by generating basically using Monte Carlo and a scoring function that scores them. how well they are placing their groups like if you want to throw any shape where two segments occupy the same space where they overlap in space, so you discard them so we can use Monte Carlo in the same way that David Baker's Rosetta program does for find shapes that can do specific things, just search with Monte Carlo and a scoring function that rates how well you are matching the shape you want to take, an example of what we are doing now is we are putting metal bonding groups in our molecules in spiritual liquors.

I have to stop saying I want everyone to use them, but right now they are the only people doing these things we put metal binding groups on them, we go to the Cambridge Crystal graph database, we get crystal structures of metal ligand complexes using the same ligands that we are incorporating into the spiral ligamers and then looking for molecules that can organize the groups to match the metal binding site that we got from the crystal structure, we can then change our metal binding site to eliminate a type of group and leave a hole and an open coordination set in the metal where other groups can enter and catalysis could occur and this is how we can begin to rationally design metal catalysts with a spirulina scaffold around to create open coordination sites and voids and pockets around those coordination sites to control what can reach the metal.

I guess I hope I answered your question. Yes, yes, definitely, so we have a little bit of a folding issue, but it's manageable. It's good, awesome. I'd love to, you know, learn a little bit so maybe you can explain to me, if you want, about any morbidity about protein binding, you know, like the application. that you mentioned here, I know you've already discussed it a little bit, but in case there are more details, if you want, that would be wonderful before we get into a little bit longer molecular topic, you'll need targets that are very quick to identify. and uh. and I think it's of a lot of interest to people in this group, yeah, so the protein binding that you should express it with, we don't know yet if these bind to proteins selectively because we're looking at all living organisms and organisms. dangerous living things than them.

I can't tell you, but we're doing this in a biosafety level 3 lab, here at Temple University, we find molecules that capture the organism in the beads and then we try to get them into diagnostic testing, but we don't. . We don't know what part of the organism we are joining, it's a little strange the way this whole project was set up. We are now talking to them about how to target a specific protein where we will know what the target is and we can get a cocrystal structure. of the protein with Simba attached to it and discover how it binds.

This is an important thing to do, but we're not doing that yet. We are working with living, dangerous and unpleasant organisms. Well, thanks for the clarification and then. You know, the last one and the bottom right one on the side that's obviously a big industry. A lot of people here are the American analytics app and you know maybe it's a little further away. I'm not sure about your perception, but what do you think we do? I could get excited about this potential, so, you know, what got me into this was Eric Drexler, you know, and the creation engines and then the systems shit, it's here, so you know, I read this stuff .

I loved it when I was a grad student, that's what I have. I got into this. I really like less what they were talking about about molecular building blocks, so positional synthesis with a few atoms, um and um, I always thought that, from the beginning, the idea would be to build these building blocks, build larger structures, functionalize them . on the outside so that they can pair with each other so that they are complementary to each other, but they also use chemistry to put them together, so they basically create molecules that are complementary so that they can couple and then have groups that will react when they come together.

Re-dock and lock, so the idea was to dock and lock and be able to first self-assemble complex structures that could be machine parts that could then be used for positional assembly. I don't think about it too much because I'm really kind of immersed in making chemistry work for the last 20 years, but that's always been the goal and I think we're close to achieving it now because we were able to synthesize, we synthesized two and a half million, five We kill Dalton macromolecules with a lot of functional groups on them, that's what you need to be able to do this.

We've scaled the building blocks up to kilograms, that's what you need to be able to do this, so that's where I really want to go with this. It is one of the instructions to follow to build a first generation molecular Nanos technology that could then do more controlled chemistry in the way that Eric talked about from the beginning. Awesome, thank you and have you ever seen that you know like in this part like maybe this more? long-term path if I see similar risks or challenges that you think we should at least take into account as we are building them as something quite sophisticated in a new structure, there are always, of course, there are always risks with any technology whatever it is It's almost a topical, um, I'm not worried about some of the early risks that Eric wrote about doing things like self-replication, it's extremely difficult, shit, just to get something off the ground, it takes a lot of work to make something that could actually be. dangerous um it would take a lot of work uh you know, but there's a lot of biology lessons about this, you know, proteins, right proteins, 20 amino acids, they're harmless, they're delicious, we eat them for breakfast, um, but you put them. the right combination and you get a botulinum toxin that is one of the most lethal molecules known, you know that there are three units, one that binds to the cell surface, another that translocates through the membrane and then this catalytic domain that cleaves these trap peptides that prevent acetylcholine.

Bags of excuses for Colleen to squirt out of the cell and prevent the cells from communicating with each other and causing her brain to spread. They are the most lethal molecules known. It's not about what you put there, but how you put it. so the right combination of Sphero ligamers yes they could be very dangerous let's not design them let's focus on things that will really help us there is a lot to do you know solar fuels diseases of aging we have so many problems we need better molecules to solve them I have been working for 20 years to develop what I think are better molecules, more programmable molecules, and the software to design them is ready to go.

I need more people, I need more resources to do this wonderful thing. That brings me to the last question because We're on time, which is the standard question: if people are now very excited about your work, which I think they are, then what can they help you in particular and your group and really advance this work? Contact me and if I don't respond be persistent, contact me if you have experience in software engineering, we have a big project we are working on and we can make unused C plus plus using llvm is our compiler. end um, I have a talk on YouTube about llvm, it can do chemistry compiler stuff um uh, if you are interested in chemistry, come talk to me if you are interested in investing in this, come talk to me, um, there is, I want tell the truth? all those things, uh, right now, unfortunately, it's just us, we're the only people making them, but I really like other people making these things and we're working hard to scale them up and reduce the cost and the complexity, okay? wonderful well I'm volunteering at least for everyone in this group to check me out in case Chris doesn't get back to you for a while and then I'll join the ping crowd and so I can I can too.

I'll be happy to open YouTube, okay, wonderful. Hey, this was really amazing and it's been crazy to also see how much progress has been made since the last workshop where we saw, you know, a partial presentation at least of this. I'm really excited to see how you're representing yourself in the upcoming workshop, you're definitely moving pretty quickly, so thank you to everyone, um to uh, for joining this one here at the seminar and to everyone on YouTube for watching, uh and Yeah. We'll be back very soon with the next um and yeah, please contact me if you can't get in touch with Chris.

Thank you very much Chris. Have a wonderful day everyone and I will see you very soon. Yes, thanks. you all