 So I would propose that, well, we have a good 40 minutes for discussion, so we already tackled a number of issues. I would start by asking the audience whether you have what you may call technical or specific questions for any of the speakers. And then once we are done with that, we can go to more general discussion around the points that have put there as a proposition. We are welcome to propose other things as well. But, you know, let's start by making sure that whatever specific technical question on the projects or the work that has been reported today and that we may have is raised. So, any questions for any of the speakers? Yes, go ahead. Okay, so you claim that more letters means faster evolution. It is partially, I understand that this is not partially a joke partially, but it's a very serious question. Because more complexity means more flexibility once, but higher dimensions are a longer optimization. Well, do you have any idea what is the optimal number of letters for faster evolution? Seven. No idea. You know it's four. Yeah, actually, I think four. Me too, me too. Seven. Okay, these are questions that I would like to return later in the next session. Any other specific technicalities? Yeah, all right. Yes, I would like to ask Morgan about the self-made systems that he is investing in so much, because it's very interesting. So, you mentioned that the Sylvus Megaterium set the extract. So, what are the advantages of the kind of protein yields? I think our strategy was to essentially take a lot of organisms that we think have utility. So, Megaterium has been proposed as a potential organism for industrial production of proteins. And I guess we just wanted to have some sort of comparator that we could compare, you know, E.coli, the Sylvus Megaterium, the Sylvus Subterus, pseudomonas putata that our chairman is very interested in, and other types of organisms. And then just try and get some understanding about is there any correlation between how they, you know, the organism or phenotype, if you like, and how the extract behaves. It's a really ridiculous question, I realise. But it's just a sort of interesting question, you know, and so if we can do it really quantitatively, we might find some interesting differences. So, Megaterium does seem to have, so it could be, you know, the number of ribosomes, it could be the efficiencies of some of the translation processes between... Who knows? I don't know what the answer is, but the idea is to try and explore that in a more systematic way. But, you know, typically, you right now see that it behaves better than the other expects, or it's just... It certainly starts to produce more of our standard-prating GFP, yes, that is correct. You know, under the same, you know, in the same conditions, in the same volumes, etc. But there is one thing I have to mention is that cell-free extracts are quite heterogeneous, and you need to optimise them for some quite strange parameters. One is the potassium ion concentration, as well as the magnesium ion concentration. So there's obviously some salvation going on, there's some strange behaviour within the extract. I'm talking about the function of the reader, this transcription translation, that's our reader. And for that to be effective, you have to optimise the extract using these particular ions. I'm not clear why, it'd be very interesting, and it's clearly something to do with salvation, but we haven't got... But ATP needs to be used, right? It varies from extract to extract. How do you break your cells off? Is it like nutrition-based, or mechanical? Yeah, mechanical. We'd usually use sonication. Yeah, but sorry, the optimisation seems to vary from extract to extract from organism to organism. And even from batch to batch, so obviously there's... When we're working there. I'm especially curious, now that you start to look after therapy, will you test your monomers for cell toxicity? Yeah, which I couldn't. We have just done cell toxicity. In this case, it's not so big toxicity. I had a question for Paul for your Turbo Sigma 70 promoter that doesn't perform in vitro, but not in vivo. Is this related to RNA levels or protein levels? Did you check in vitro if it makes...? Yeah, it's a good question. We're doing it now. I'm sorry, I can't answer that. The only thing I can say about these series of synthetic promoters we made, there are very small differences between them, and this one is a small difference. So we'd be very interested to know exactly the question you've raised, and also the context issue as well. But sorry, I haven't got the data yet. Just following on from that, have you thought about using that promoter in vivo as a readout and then doing the mutagenesis to find out what's holding it back? That's a good question. No, we haven't. That's a very good idea. We should do that. That's interesting. Thank you. Okay, great. So once we're done with these technicalities, I would like to invite you, and not just to raise questions, but also to actively participate in this more real discussion that is connected to some of the topics that have been raised by the speakers and by myself. So let me just fire by raising the issue of standards. We have heard about the importance of standards. Standards have been one of the tendencies of the biology through ND, IGN, all these things. But the standards are at the same time. When you talk about it to our colleagues, at least as my experience, people don't get precisely very, very excited about that. So everyone recognizes that it's an important thing, but no one is really very willing to put effort and money into developing with standards. So that's one aspect. The other aspect is that if you look at the history of standards, and I was looking before, standards in many cases have been adopted by what you may call bottom-up initiatives, but in many cases they have been implemented by brute force, and I think that the case of Napoleon is a good one. So the reason why we enjoy the metric system in the continent, and partially in the UK as well, is because of Napoleon, you know, because before that it was a whole chaos of standards for measurement and establishment. I mean, how do you see this topic developing? I mean, do you find that this is really important? When I was listening to you, I was thinking, well, maybe you don't care at all. You don't care at all about the standards. You do things and you ignore whatever standard may be involved with that, and you are actually happy with that. In other cases, people may claim that standards are bottlenecks. So unless we have a common standard for doing things, there are some areas of biology and things that we'll never develop. So what is your sentiment? What is your feeling about that? So maybe Paul can start advocating standards, and maybe others may disagree. No, I'm a great proponent of technical standards. I mean, the thing we've got to differentiate, I think, is a little bit between standards, as in the ISO, International Standard Organisations Certification, and the sort of thing I think we need now, which is more a greed set of technical standards among the community of how you measure things, actually. And I would pin it down onto that metrology. So I just think it's, you know, when we measure something in one lab and we measure it in another lab, can we ensure that the measurements are comparable? And even in biology, I think we don't really push that envelope too much. We're better than we were, but I think, you know, so for example, if someone's doing a fax analysis, I'd like to know what the gates are, what the whole metadata is associated with the measurement. Very tedious stuff. A lot of biologists are interested in that. But I think if we're going to get this kind of quantitative comparative sense of measurement and standard processes and standard operating processes, then we're going to have to push that forward. You know, assuming we want to make this field a kind of engineering type field, because I can't see how it can progress without that. I don't think we could have an engineering field without standards, because of me. Yes, Eric. Yeah, so I've been involved with generalisation for a long time. So I have several things to say there, but I just would like to say so. Measurement is obviously one domain that is extremely important, but that's kind of general science. You want to have proper visibility in your papers and so on. I think what we have to look for is not standardisation as a purpose in itself, obviously, which has sometimes happened a little bit, but really try to find those points where we want to exchange things. I mean, I think a standard is something like an interface where you start doing something, and I can then take it and continue it, and then somebody else can again continue those crossover points. Those are the ones that we have to identify, and this is what we should standardise right now. I think measurement is one of those. The exchange of designs is another one. The information that goes into the paper that we publish is still, usually it's still like a really big pain to actually identify the sequences, the DNA sequences that go with all those nice synthetic biology papers, and I think that's a little bit embarrassing, and so that's something that the S-ball community where I'm involved is working on, but also the S-ball community kind of lost track of that a little bit because they then focus a lot on systems biology modelling and the higher level, but the very basic level, I think this is what we should go for, like basic problems like I have the paper, I have five or ten or five hundred plasmids that are described there, and we have to make sure that I get this design out again and can use it, and then I make a measurement and I want to make sure that I can actually work with this machine. So I think those are the things we should focus on. Okay, so let's have a brief introduction, so we have a more dynamic discussion. Thank you. One more comment and then we move to the next topic. I agree and I disagree with Philippe, but if I want to understand biology, I am against the standard. It's a way to block the innovation and I think the lighting organism is exactly what you said that exists today. It's because they are not standardised at all. If from an engineer point of view, I want to propose a product to the population or if I want to make something, I need standard. So if I am a scientist, I don't want standard, I don't want to be blocked by any kind of standardisation. If I am an engineer, I want standard. Okay, so I think that you can see the entire landscape of opinions. I think they are all very good. My own take is that we should concentrate for a time being in metrology to really have good standards for measurement of things that is very weak in biology. So yes, one second. It's not a comment, it's a question of a report. Do you say that this problem of establishing standards is a technical problem because it's very difficult, it's a human problem, or is it more a conceptual problem? You have to understand deeper to your field to be able to establish standards. So there is a link between what this person said. It's too early for him and it can come together. I think the word standards is off-putting, Victor said, and maybe we should not... standards means something to everyone differently. To me, standards, I'm talking about technical standards, reproducibility, standard operating processes. That's all I'm talking about. So I know that if I follow exactly everything you did in that paper, I'm going to be able to reproduce that measurement. And to be truthful in a lot of biological research, that is not the case now. We've got to move from that. Of course it's a human thing because it means I've got to persuade my post-ops and students to put down all that ridiculous extra bit of data called metadata about the measurement. I've got to replicate multiple times in the different conditions. It's really tough. So that's why automation, I think, is the way forward. So a lot of our labs are developing automatic platforms for characterizing promoters and parts so we can have throughput automation and standard measurement. I'm talking about the engineering of biology, so it's not about blue sky biology because that's different. That's completely unencumbered by anything you want. You just got to do what you want. But if we're going to build a synthetic biology that has applications and it's going to produce a new kind of way of doing biotechnology, then I think we have to have it. And it hasn't happened today. That's my case. We have one more question. Any other discussion? I just want to say, I think standardization is very difficult. We should know that. So when we just determine the DNA or protein concentration, it's still very difficult. Do you know? We know. All right, so I would like to take the next point of discussion about this modularity that has been raised first by Paul, but then to an extent has been raised also later. So what do you think? And this is a question for the whole audience. Is biology modular as it is? Can it be made modular? And if we manage to make it modular, will it stay modular? Questions, opinions about that? Or maybe you never thought of that. Yes. I think we have to talk about modules because if we are able to define modules, if they exist, then we can also think about standards, how to link the modules. Otherwise, the standardization makes not a lot of sense. And if we talk about modules, and if I look at systems biology, are we able to separate organisms in modules and describe them by sets of equations and link them? I think we have not been very successful. All right. So the motivation for this question is that you look at the textbooks and then you see, you know, metabolism. You have this metabolic block, you have the glycolysis, you have the pentose pathway, whatever. And then you have, you know, DNA binding proteins. But at the end, you know, there's this growing evidence that the same protein may make at the same time many, many different things. Like some proteins involved in glycolysis may, at the same time, bind DNA or some protein that makes interaction at the same time is a structural component of this other thing. So even though our brain may be programmed to see modules all over the place and to see divisions, maybe this is something that does not happen in reality. And I think that you may agree with me that every time we want to have a perfect model in biology, sooner or later we'll run into problems. But maybe there's some hope perhaps through xenobiology or perhaps through physical containment or with other strategies to remake a model of reality. Why do you think about that? Do you think that this is our way to go? Yes. So I think that as a biologist, when thinking about modularity, you need to consider whether the modules are insulated from the system or whether they're part of the network as you were getting at. So if they are, for example, part of primary metabolism, they're likely to be pretty well networked. Whereas if they are specialized metabolic pathways that are lineage-specific, they may be sequestered away with less... So it's the connections between the components of the module and the rest of, let's say, metabolism. So if there aren't any, there are very few, you have to have your precursors. But if the specialized part is insulated, then that may be easier to deal with in a living organism than the whole of primary metabolism. All right. More opinions? Yes. So I guess there's another term that people have used which is orthogonality, right? Which gets to this idea which I think Paul mentioned too, which is whether what you design does not impact the original system when you introduce it. Though I'm not entirely sure that's necessarily the modularity we're discussing with the arts, but in terms of the systems, whether the design component has no effect on the non-designed component, which is the organism itself that we so are. Very good. Mark, you want to share your thoughts with us? Do you want to share your thoughts with us? I'm just saying, this is a flow of information that you can see from part to part. If these different questionations, this network, we can call the model if not using a model, right? You know that, say, in the brain, there are models that have connections in certain regions, much denser than some others. In reality, you raise the problem of orthogonality and orthogonality along with the standards and other magic words are one of the preferred terms in synthetic biology. Is orthogonality a realistic objective or we will always run into different bits and parts? I think modularity is a reasonable goal and the only way to get somewhere is to start addressing it and trying to make things modular because we're going to learn things that we didn't previously see when we weren't looking for this goal of isolation or orthogonality. But the counter to that is if we make things completely modular and remove orthogonality, remove those networking connections, we're also going to inherently become even more complex and less efficient. So, after we build in that modularity, we're going to have to build back in efficiencies once we completely understand if we can ever get there. So, there's sort of a... I think the reason nature is not so modular is because it has to be more efficient than what we're aiming for right now. But you argue that one of the ways to get into this modularity would be to use your approach to really isolate physically whatever biological components that you have at hand. That's my personal goal, yes, is to isolate it in that way. Yes. I'm sorry, I came back because the example of CAPSI is an example of modularity. So, you have a face. You have 150 faces and you connect with the same brick, one brick only, you connect everything and you do a complex object. So, this is an example of modularity. My question before is to know if this modularity is imprinted in DNA or not. But is it a p-genetic phenomenon? That is my main question. I think we like to believe that it's completely imprinted in DNA, but it's also controlled by many other things like you were saying, the salt concentrations or... So, the information is there, but we also need environmental conditions tonight. Good evening, Rick. Yeah, I think this discussion comes up a lot. Basically, nature is in... I mean, biology is incredibly modular, I think, because otherwise none of this genetic engineering that we have been doing in 30 years would have ever worked by taking a gene from one organism, putting it into another organism and actually having a product out. So, it's clear that at some level there's an amazing amount of modularity because we can do that. It's not a perfect modularity, but even a computer that we have engineered is very modular, because all those circuits on the chip, even though they're very modular in the layout, they all depend on each other with the MPS or with the current that they're using and so on. So, I mean, even in perfectly engineered modular human systems, this modularity is never perfect. So, obviously, in biology it's even less perfect because there's, as you said, there's this tendency to integrate things, to make them more efficient, but then evolution tends to break them apart again, and that's exactly what we're looking at. So, it's a continuum between it's definitely not a perfectly integrated system, it's definitely not a perfectly modular system, but it's modular enough to have a starting point. Okay, so I look at the clock and we have just a few minutes left, and I would like to finish without raising what I find a very, very interesting discussion on genetic alphabets. And that was initiated before, but I would like to get more feedback from you and from the speakers. I mean, how far we can go in developing new alphabets in the sense of developing new functions and new life forms and so on, or it happens that evolution has found that kind of optimum, and that's it. So... Wait, wait, wait, let me... Okay, so the idea is, I mean, by having, you know, the Phoenician alphabet has 20 few letters, the Latin alphabet has 28, and Chinese alphabet has many more. Okay, whatever. So... The question is, by having more letters, can we write different stories? To me, what is important? You know, because maybe the option is to have a small number of letters, and with that, to make a combination of them, and with that, you have enough to write all types of histories from the Iliad all the way to the latest Nova. So by having a different alphabet, you can write different things. All right. All right. Wait, wait, wait, wait. So those that are more engaged in this type of developing an alphabet, can you say your opinions? No. You can do Iliad with zeros and ones. Can we hear from those that have developed new letters for the Iliad alphabet? I'll defer to a chair. Okay, please. No, no, because... No, no, it's not. You have to say something. But... Yeah, through my experiments, so on Earth, four base, 20 aminarsity is the best one. Well, 23, you know. Yeah, yeah, that's right. But so it depends on the, maybe, fidelity. So that is very important. So if we make a DNA, so replication, in this case, so 20 aminarsity polymersity is important. And in this case, maybe the fidelity is a present fidelity is the best one. So that's why our letters are limited for. So if we add aminarsity more than ready, so maybe we can much more try a fidelity as polymersity. So that we can improve. All right. I have a question. You have a new alphabet. You make code with this alphabet a huge amount of information. You produce something like a DNA system, but it lives in an environment. Therefore, it has. You cannot take it off from that because you have to transfer this information loop. You have to transfer it in the reaction of the environment. And therefore you couldn't use any chemicals or substances putting up an alphabet. You had before what kind of pieces of alphabet, for your alphabet, was it parts of DNA, or what was it? What were the substances you used for your system? The precursors that give rise to the altered base, the chemical precursors. What did you use? I mean this must be compatible with the environment. Otherwise it doesn't work. So our natural basis can make forms of nature so just completely man-made. So that's different from natural basis. You had a natural basis. Natural basis it's astrobiology. Actually I don't know, but natural basis can make on us or from the universe. Alright. A couple of comments. Real quick question. Maybe someone should run the experiment and see if they can evolve better after this. Exactly. These questions are addressable experimentally. Organisms with additional amino acids or with additional bases or challenged amino acids or whatever. They could be put in competition with wild type species or whatever. Another experiment which should be very something restricted instruction set computers. So that you try to restrict and modularize and so on. It will be very interesting to try to attempt to make organisms with an amino acid less, not more, less to see whether and put to the test the notion that the more the bigger the alphabet the more evolutionary and adaptive options. We believe so, but it is not an experimentation. Yeah, indeed. I think these are fascinating questions and also the other question of how to reduce the alphabet and what happens when you reduce it. And I know that some of you are working on that. Okay, ladies and gentlemen, I think that we have just enunciated. Is there any super... I just wanted to say it depends what you want to do with the alphabet. If you want to include information zero and zero and one, so it's enough principle, it's just a string that's very long. But if you want functions if you want these things to fold up into three-dimensional shapes, I think it's pretty clear there's also classic experiments from Jerry Joyce's lab. That if you reduce the alphabet from four to three to one functionality drops and Mishir Israel suggests, certainly there's a first glimpse that if you increase the alphabet functionality, the functional potential the adaptability goes up. So one has to distinguish between that and the other. Great. Okay, ladies and gentlemen, I'm after you have to maintain the program in time or less to have some time now for refreshment, but I would like to close without thanking the speakers.