1 00:00:02,490 --> 00:00:06,570 So possible. Can you just tell me your name and your current affiliations? 2 00:00:07,760 --> 00:00:18,120 I'm Annette. Yvonne Bell and I work at the University of Oxford as a translational scientist for the Nuffield Department of Medicine. 3 00:00:18,120 --> 00:00:21,870 And the nephew of Department of Medicine is very big and very large. 4 00:00:22,200 --> 00:00:29,250 So it's a a institute within the nephew of Department of Medicine that is called Centre for Medicine's Discoveries. 5 00:00:30,330 --> 00:00:36,270 That's great. Thanks very much. And without telling me your entire life story, but just summarising briefly, 6 00:00:36,270 --> 00:00:42,510 can you tell me how you got interested in medical science in the first place and how you got from there to where you are now? 7 00:00:44,520 --> 00:00:52,410 It was quite a journey to arrive where I am right now and currently I'm working in drug discovery. 8 00:00:52,440 --> 00:00:54,870 But this was what I've always been doing. 9 00:00:55,230 --> 00:01:05,390 I am originally South African and then my parents moved to Germany at some point and I studied medicine in Germany and. 10 00:01:06,550 --> 00:01:12,700 At that point, I wasn't quite sure whether this was what I was interested in doing. 11 00:01:13,030 --> 00:01:18,669 I also studied music on the side, which I tremendously enjoyed. 12 00:01:18,670 --> 00:01:29,649 But then after I finished my medical school, I got offered a Ph.D. at Oxford in clinical medicine and where I worked on vaccine development. 13 00:01:29,650 --> 00:01:30,430 And interestingly, 14 00:01:30,430 --> 00:01:43,480 actually on the vector that played a role in and I worked on hepatitis C and immune responses and then on vaccine design for hepatitis C in £80 group. 15 00:01:44,220 --> 00:01:50,790 And that was really the point where I got really interested in infectious diseases overall. 16 00:01:51,460 --> 00:01:55,900 I, after finishing my PhD, decided to go back to clinical medicine. 17 00:01:55,900 --> 00:02:03,160 So I had a couple of years of clinical work where I worked mainly as I rotated through the usual 18 00:02:03,160 --> 00:02:11,350 foundation program and ended up working in sexual health and in HIV clinics that I really enjoyed, 19 00:02:11,830 --> 00:02:15,640 and mainly because they were around infectious disease and viral diseases. 20 00:02:16,690 --> 00:02:20,979 And then for family reasons at some point decided to have a break from the clinical training 21 00:02:20,980 --> 00:02:27,300 program where I had a gastroenterology post office and kind of go back to research again. 22 00:02:27,310 --> 00:02:40,180 That was in 2018. So I for about two years worked at 60% for the university and a research position and then addition in addition to that. 23 00:02:40,630 --> 00:02:45,520 And in the hospital, I'm doing sexual health and HIV clinics. 24 00:02:46,180 --> 00:02:49,730 And that at that point seemed to work quite well. Well, for us. 25 00:02:49,750 --> 00:02:56,739 So for me for me personally, in terms of interest, but also for for us as a family, as a as a as a good set up. 26 00:02:56,740 --> 00:02:57,309 At that point, 27 00:02:57,310 --> 00:03:06,670 I had like three small kids and wanted to have a little bit more flexibility in what shifts I was doing and how many shifts I was doing at the time. 28 00:03:06,670 --> 00:03:12,400 So that was a very good set up. And then the pandemic hit. 29 00:03:12,970 --> 00:03:15,790 So that's probably where your next question was. 30 00:03:15,940 --> 00:03:25,480 Well, it's not quite my next question, because I want to explore a little bit more of your your expertise in drug discovery at the moment. 31 00:03:25,840 --> 00:03:28,100 But first of all, there's a question I'm just asking everybody, 32 00:03:28,150 --> 00:03:33,930 which is which what if there was one kind of big question that really drives you in your research? 33 00:03:33,940 --> 00:03:40,030 What would you say was. I think. 34 00:03:41,090 --> 00:03:48,440 What what really interests me is how we can get from a very basic research finding 35 00:03:48,890 --> 00:03:56,780 to a patient and how we can do that in a way that enables whatever we come up with, 36 00:03:56,780 --> 00:04:09,800 whether that's a vaccine or whether that's a drug or whether that's a technology to be accessible to populations worldwide and also relevant. 37 00:04:10,040 --> 00:04:19,730 So how can we make sure that we don't waste our time on something that is actually not going to be used or not going to be accessible in the end? 38 00:04:21,170 --> 00:04:26,959 So it's it's quite, I suppose, quite multi-disciplinary what we're interested in doing. 39 00:04:26,960 --> 00:04:31,340 What, what, what would you say that the main methods and techniques that you use in your research. 40 00:04:34,490 --> 00:04:41,990 So it is quite difficult to nail it down to kind of different methods because it's a very broad field. 41 00:04:41,990 --> 00:04:48,770 Like I've been working in vaccines previously and when it came to vaccines the question 42 00:04:48,770 --> 00:04:54,380 was how do we have to design them to make them relevant to the circulating viruses? 43 00:04:55,550 --> 00:04:58,880 For for breakfast? CURRY The question is actually quite different. 44 00:04:59,240 --> 00:05:08,090 And we're using very different research techniques and different translational techniques to get this drug to the clinic. 45 00:05:09,550 --> 00:05:15,590 But here the of the one big question that I'm currently working on with a very, 46 00:05:15,590 --> 00:05:23,300 very widespread team that works in lots of different institutions and how we can generate the drug that's not working. 47 00:05:23,760 --> 00:05:34,220 And that's a kind of a compound that is basically doing its job and safe enough and all those kinds of questions. 48 00:05:34,490 --> 00:05:43,880 But also, how can we make sure that we can get the compound to the patient in time, in different locations, 49 00:05:44,870 --> 00:05:52,310 and how can we set up and licensing the patents and everything around that compound that. 50 00:05:54,590 --> 00:06:00,049 It is actually a compound that can be given and that is not too expensive and 51 00:06:00,050 --> 00:06:06,590 that is accessible in Bangladesh as well as in Rwanda as well as in Brazil. 52 00:06:07,580 --> 00:06:15,530 The key point so it pre-empts my next question a bit, which is why can't we just label this drug companies is not the. 53 00:06:18,250 --> 00:06:19,720 I mean, traditionally. 54 00:06:22,280 --> 00:06:32,960 Drug discovery has been made in big pharmaceutical companies, and they do have and big pharmaceutical companies come with a tremendous network. 55 00:06:33,290 --> 00:06:39,260 And what they have and that can't be underestimated is an enormous logistical experience. 56 00:06:41,200 --> 00:06:48,880 But on the other hand, all the drug companies have to pay their employees and they run for profit. 57 00:06:49,360 --> 00:07:00,790 And and and they they are mechanisms in place to make sure that patients and in the global south do have access to the drugs. 58 00:07:00,800 --> 00:07:04,390 But all these mechanisms are not quite perfect. 59 00:07:04,420 --> 00:07:13,480 So, for example, the medicines patent pool can provide drugs at lower prices to to low income countries. 60 00:07:13,720 --> 00:07:19,300 But there are some low and middle income countries that are left out from licenses from the medicines patent pool. 61 00:07:19,840 --> 00:07:27,610 And it's the question that I'm working very closely with, and that's like Neglected Disease Initiative at the moment, 62 00:07:27,610 --> 00:07:35,440 is and how can we potentially enable these patient cohorts that are in, 63 00:07:35,950 --> 00:07:43,149 say, medium income countries that may not have access to the very high priced drugs on patents, 64 00:07:43,150 --> 00:07:46,870 but neither have access to drugs on the medicines, patent pool drugs. 65 00:07:47,170 --> 00:07:51,190 How can we enable those to have access to the grants, for example? 66 00:07:51,760 --> 00:08:00,260 And it's. It's not directly linked to the actual research that I'm doing, but what we are trying. 67 00:08:01,210 --> 00:08:10,450 This was a unique experiment that we really run of the pandemic, which was Can we go all the way into clinic for open science? 68 00:08:10,450 --> 00:08:15,040 And we haven't answered that question yet. We're still in the process of doing it. 69 00:08:15,050 --> 00:08:21,640 We've got compounds in pre-clinical development at the moment that are not patent protected, 70 00:08:23,240 --> 00:08:30,010 and we're trying to work out whether there is a way to set up and enough buy in from generic 71 00:08:30,010 --> 00:08:38,640 companies and buy in from governments and buy in from access groups to make these drugs available. 72 00:08:38,680 --> 00:08:47,190 So that's what that's the kind of if you want a more social science problem that is unrelated to actually making a safe and efficient drug. 73 00:08:47,770 --> 00:08:51,310 So you use two terms that which I'd like to spend a little bit more time on. 74 00:08:51,730 --> 00:08:56,830 One is neglected diseases. One, what are neglected diseases and why are they neglected? 75 00:08:57,700 --> 00:09:03,489 So neglected diseases is quite a wide term and I'm covered. 76 00:09:03,490 --> 00:09:08,770 This definitely not one of the neglected diseases of the moment in terms of research, 77 00:09:09,220 --> 00:09:19,000 but I think and traditionally neglected diseases is a class of disease where not a lot of research, 78 00:09:19,000 --> 00:09:31,360 time and effort within pharmaceutical companies has been spent on generating and generating compounds or drugs against those diseases. 79 00:09:31,630 --> 00:09:36,370 So are diseases without a lot of treatment options and diseases where there is 80 00:09:36,370 --> 00:09:41,319 not a lot of translational research ongoing to generate those treatment options. 81 00:09:41,320 --> 00:09:51,399 And and for example, the reason that Drugs for Neglected Disease Initiative in Switzerland exists is to look at exactly these diseases 82 00:09:51,400 --> 00:09:59,950 and make drugs against those that help patients that have these kinds of diseases to actually get treated. 83 00:10:00,400 --> 00:10:06,400 And leishmaniasis is one of those Chagas disease is one of those and then a couple of other diseases. 84 00:10:06,730 --> 00:10:10,660 And I wouldn't say that covered the such pools into this category, 85 00:10:11,410 --> 00:10:17,950 but and neglected diseases can also kind of line up with the neglected patient cohort. 86 00:10:18,340 --> 00:10:21,640 So kind of hold that just doesn't have access to disease. 87 00:10:21,880 --> 00:10:26,800 So for example, hepatitis C virus is not a neglected disease in any way. 88 00:10:26,890 --> 00:10:37,120 They are drugs available and millions of billions of pounds have been spent on on hepatitis C research in the last 20 years. 89 00:10:37,510 --> 00:10:40,540 There are plenty of direct acting antivirals, 90 00:10:41,350 --> 00:10:48,520 but there are still neglected patient cohorts that don't have access to this kind of treatment at the moment, 91 00:10:49,060 --> 00:10:57,700 even though they have been around for a long time. And and actually some of the problem has been obvious for HIV for a very long time. 92 00:10:58,000 --> 00:11:04,600 And you can even go a step further and say they are neglected patient cohorts that don't have access to insulin and diabetes, 93 00:11:04,840 --> 00:11:11,440 like diabetes patients that don't have access to insulin, which is a drug that is off patent and should be available to everyone. 94 00:11:11,800 --> 00:11:19,880 So it's an. And neglected disease doesn't necessarily equal a neglected patient cohort. 95 00:11:20,240 --> 00:11:24,260 And not all of the questions have to be answered with novo discovery. 96 00:11:24,380 --> 00:11:30,050 A lot of the questions can be answered with making drugs available to certain patient cohorts. 97 00:11:30,960 --> 00:11:38,510 And this is not my area of expertise, and I'm very, very grateful to be working with the FBI on this problem, 98 00:11:38,510 --> 00:11:46,160 because they have the kind of IP lawyers, access specialists that are looking into these questions. 99 00:11:46,550 --> 00:11:52,790 The one unique thing that the moonshot, the kind of the research effort I'm involved in, 100 00:11:53,420 --> 00:11:57,750 can contribute to this theoretical question of how do we make drugs? 101 00:11:58,310 --> 00:12:02,150 The sensible is that it is an open access effort. 102 00:12:02,330 --> 00:12:10,670 Right. So that brings me to my next question. So I still haven't quite got sober yet, which is about you talked about open science, open access. 103 00:12:11,120 --> 00:12:12,259 What do you mean by that? 104 00:12:12,260 --> 00:12:22,730 And what precedents have there been within Oxford for setting up conditions under which that kind of open approach can take place? 105 00:12:23,630 --> 00:12:28,820 So I am talking about SGC, just. Yes, I am. 106 00:12:31,330 --> 00:12:40,210 Open drug discovery again. If you ask ten people would get ten, ten different opinions on what open drug discovery actually means. 107 00:12:40,730 --> 00:12:48,500 And so being SGC and Oxford has that field. 108 00:12:48,520 --> 00:12:59,679 I should have said the study consortium has been founded in 2004 and actually doesn't only only have an Oxford site. 109 00:12:59,680 --> 00:13:03,850 It was a worldwide effort and it has a Canadian side. 110 00:13:03,850 --> 00:13:08,860 And Edwards has been absolutely key to this setting up this effort. 111 00:13:08,860 --> 00:13:18,130 It was very much his personal interest in looking into how can we generate. 112 00:13:19,860 --> 00:13:34,079 Only tools that enable drug discovery in an open way across companies that are how can we kind of basically pool efforts and across pharmaceutical 113 00:13:34,080 --> 00:13:42,000 companies so that if the idea is actually quite revolutionary and it's getting different pharmaceutical companies to pay into a covered pot, 114 00:13:42,540 --> 00:13:50,069 they can nominate targets. And then there are different sites worldwide that bring different kinds of expertise. 115 00:13:50,070 --> 00:14:01,110 And they will work on enabling these targets by developing tools and by tools, I mean generating structural data. 116 00:14:02,250 --> 00:14:06,270 So this was the very first step. So that's why it's called Structural Dynamics Consortium. 117 00:14:06,420 --> 00:14:11,510 So they would take some genomic information. This was specifically for human proteins. 118 00:14:11,510 --> 00:14:17,370 So no viral proteins involved at that point. This is just specifically to target the human genome. 119 00:14:17,370 --> 00:14:23,580 So they would go for a whole class of human proteins that at that point was under investigated. 120 00:14:23,910 --> 00:14:33,180 So you wouldn't have this structure available and then kind of work on a whole class of proteins as a kind of wholesome approach. 121 00:14:33,450 --> 00:14:40,470 Because what you also have to know is that if you saw one protein structure that doesn't yet gives you very much information, 122 00:14:40,740 --> 00:14:46,980 you will want to know what the difference between this structure is and all the others to make sure 123 00:14:47,190 --> 00:14:54,600 that you can develop a drug specific against the one portion of the class that you are interested in. 124 00:14:54,990 --> 00:15:03,480 And there were actually several other grants that took this concept a step further that were also at the SGC, 125 00:15:03,660 --> 00:15:10,139 where they didn't only look at the structure such, but also developed an acid for protein. 126 00:15:10,140 --> 00:15:17,310 So this basically is is a biochemistry tool that you use to show how your protein works. 127 00:15:17,850 --> 00:15:22,559 So you basically, if you, for example, as we will be talking about proteases later on, 128 00:15:22,560 --> 00:15:29,870 these proteins that chop things up and keep other proteins or and and in. 129 00:15:31,280 --> 00:15:43,339 And and material. And and these proteases and all the functions of these proteases you can assess in an essay and an assay, 130 00:15:43,340 --> 00:15:47,930 we'll have a readout that you can measure at the end. So you would look at, say, 131 00:15:47,930 --> 00:15:58,010 a little piece of protein and that and you will find some way of showing whether it's together or all cleaved into two pieces. 132 00:15:58,310 --> 00:16:00,290 And that's a function of your protease. 133 00:16:00,290 --> 00:16:08,209 And then if you add certain inhibitors that are supposed to inhibit your protease, you'll see that you protein even stays in one piece. 134 00:16:08,210 --> 00:16:14,810 If you inhibitors working or it gets do two pieces give you inhibitor is not working. 135 00:16:14,900 --> 00:16:26,270 So it's a tool to kind of find out which inhibitors or which drugs or which fragments may be having some biological effect on that protein or not. 136 00:16:26,810 --> 00:16:30,799 So and this you don't develop for you one protein of interest, 137 00:16:30,800 --> 00:16:37,490 but you develop a whole protein family to kind of make sure that you get a kind of good picture across. 138 00:16:37,910 --> 00:16:44,000 And and that is really what the pharmaceutical companies are very interested and what 139 00:16:44,000 --> 00:16:48,830 the unique thing about the SGC was that they predefined but this is pre competitive. 140 00:16:49,670 --> 00:16:57,260 We can use this information and it can be used across different pharmaceutical companies and will make this open be available so that 141 00:16:57,260 --> 00:17:10,010 anyone can use this information so you can order the constructs from the SGC and and the SGC even went one step further and said, 142 00:17:10,370 --> 00:17:12,140 we're going to develop proteins. 143 00:17:12,470 --> 00:17:20,150 So small molecules that are not drugs that part earlier than drugs, but they are things that you can use in cellular assays, 144 00:17:20,150 --> 00:17:28,790 things that you can use in your enzyme assays to make sure that you can show activity against that specific protein, 145 00:17:28,790 --> 00:17:30,800 but not all the other proteins of that company. 146 00:17:31,700 --> 00:17:42,260 So that's and yes, you see has been very successful in exploring the epigenetic space, kind of different classes of proteins. 147 00:17:42,740 --> 00:17:49,580 And that and based on these probes and structures that were developed through this set up, 148 00:17:50,150 --> 00:17:59,720 there are a whole class of drugs that are now in the clinic and that that kind of bridge from original efforts that are open access. 149 00:18:00,650 --> 00:18:04,070 So I suppose yes, the actual outcomes are important, 150 00:18:04,340 --> 00:18:11,840 but the principle of that sharing of of data at an early stage in the drug discovery process is 151 00:18:11,840 --> 00:18:18,290 one that clearly has been shown to be beneficial even even to the commercially oriented company. 152 00:18:19,630 --> 00:18:28,720 There's one exception here. And what usually happens within this effort is that the compound, 153 00:18:28,810 --> 00:18:35,530 so the SGC would partner with the company on the site and target a lot of the 154 00:18:35,530 --> 00:18:41,170 medicinal chemistry efforts to actually making this probe was run within the company. 155 00:18:41,950 --> 00:18:52,360 And and one of the probes that was active at the end would be published and accessible as an open access resource. 156 00:18:53,260 --> 00:19:01,900 As part of the kind of interesting setup that a lot of the other compounds that were developed in this process might not necessarily be open. 157 00:19:02,770 --> 00:19:10,020 So there was there was an agreement within the SGC that the company can use the tools that are open, 158 00:19:10,150 --> 00:19:14,380 available to run their own discovery efforts if they choose to do so. 159 00:19:14,890 --> 00:19:19,090 And that was was done for some some companies and some SGC partners. 160 00:19:19,450 --> 00:19:23,019 And this is where the coordination is very different. 161 00:19:23,020 --> 00:19:26,780 So we'll get that. So let's let's get on to get the answer to. 162 00:19:27,100 --> 00:19:36,640 So when just first of all, where can you remember how you first heard about the pandemic in China and when you 163 00:19:36,880 --> 00:19:41,940 started to realise that this was something that we might send your own message and this. 164 00:19:42,790 --> 00:19:55,220 So I heard about. I think I first heard about the pandemic from the news, but I very clearly remember an evening in December. 165 00:19:56,450 --> 00:20:08,390 When my husband walked in and he works for the diamond source, he also told us because he works for black schools and they had a student. 166 00:20:09,340 --> 00:20:14,440 And they have an antibiotic program where they get exchange students from China 167 00:20:14,770 --> 00:20:22,360 coming to visit the UK on regular and five and kind of two monthly times I think. 168 00:20:22,840 --> 00:20:27,480 And the student had arrived in in the UK. 169 00:20:27,760 --> 00:20:30,970 May actually have been John I can't remember the date but I, 170 00:20:31,030 --> 00:20:37,419 this was kind of my first personal kind of interaction with the whole program 171 00:20:37,420 --> 00:20:43,030 and this was the student had arrived and while the student was on the flight. 172 00:20:44,450 --> 00:20:54,350 And China had started to lock down so that that student had left and came from an area where they work hundreds of cases. 173 00:20:54,740 --> 00:20:59,390 And while this thing was on the flight, they had locked down back home. 174 00:20:59,900 --> 00:21:02,719 And now the question was, what's going to happen to the students? 175 00:21:02,720 --> 00:21:11,210 Because suddenly diamond, the accommodation that the accommodation that was booked for the students said, we can't have that student anymore. 176 00:21:11,780 --> 00:21:18,620 And so within my my my partner's group, they were said to be this kind of uproar. 177 00:21:18,980 --> 00:21:22,730 What are we going to do? Like, this person has to stay somewhere. 178 00:21:22,730 --> 00:21:27,290 They can't stay in the accommodation because they're not going to have the student. 179 00:21:28,610 --> 00:21:36,739 And we were kind of discussing back and forth. And then in the end, one of the one of the researchers and friends group said, like, 180 00:21:36,740 --> 00:21:41,960 we've got an extra room in the house and my parents are happy to have her for that period of time. 181 00:21:42,410 --> 00:21:46,940 And there were no tests available. There was no way to know whether the student. 182 00:21:47,330 --> 00:21:54,379 But it basically ended up that the student that happened to find this about as a kind of self-contained 183 00:21:54,380 --> 00:21:59,420 flat where they wouldn't be in touch with anyone else until they were able to fly back. 184 00:21:59,930 --> 00:22:08,239 The student ended up staying and actually working on the kind of approaches of while they were visiting. 185 00:22:08,240 --> 00:22:13,070 But that was my kind of very strong first for them and then. 186 00:22:14,410 --> 00:22:21,850 Not a lot happened for a while. And and then we had two things pretty much happening at the same time. 187 00:22:22,630 --> 00:22:26,910 One was I was starting to cough and that was in February. 188 00:22:26,920 --> 00:22:30,909 So I have picked up COVID in February already. 189 00:22:30,910 --> 00:22:38,200 So prior to lockdown or maybe early March and in hospital in one of my clinics, 190 00:22:38,200 --> 00:22:46,269 I must've just gotten it and and started coughing and was aware that actually I actually may have picked this up, 191 00:22:46,270 --> 00:22:57,460 even though officially we didn't have any COVID get in the country and at the same time friends group at diamond and was contacted by. 192 00:22:58,540 --> 00:23:05,890 The Chinese research group that was working on the cells called the two main protein structure and 193 00:23:06,160 --> 00:23:13,870 because thanks research group at diamond can very uniquely run very high throughput screens very fast. 194 00:23:14,230 --> 00:23:21,490 And the group in China had solved the structure of the improprieties because they have been working on cells previously. 195 00:23:21,850 --> 00:23:35,160 And that's a bit more about this mpro protease. And that the the protease is an interesting protein because the kind of research community, 196 00:23:35,170 --> 00:23:41,650 similarly aspirin polymerase knows that it's a drug target that you can develop inhibitors against. 197 00:23:42,920 --> 00:23:48,860 And so right at the start of the pandemic, there was obviously the question, how can we make a vaccine? 198 00:23:48,950 --> 00:23:53,780 That was that was kind of one of the key questions and move very fast ahead on that. 199 00:23:54,170 --> 00:23:59,050 And then if you think about, okay, what else can we use in our toolbox against COVID? 200 00:23:59,450 --> 00:24:08,270 It's not only vaccines, it will be small molecule inhibitors as well, because there will always be limits to what we can do with vaccination. 201 00:24:08,480 --> 00:24:14,420 And we will, especially at the start where we have got the vaccine, yet we hopefully should have something in hand. 202 00:24:14,420 --> 00:24:17,309 And the obvious thing is repurposing efforts. 203 00:24:17,310 --> 00:24:25,549 So using a drug that is already in the clinic and that also works against the new testings and unfortunately for COVID, 204 00:24:25,550 --> 00:24:27,800 that was very, very unsuccessful. 205 00:24:27,920 --> 00:24:36,709 And in a lot of ways, I mean, we've been able to identify immunosuppressive drugs that deal with kind of the legacies. 206 00:24:36,710 --> 00:24:41,150 And I'm sure you'll be talking to a lot of people that have been involved with this kind 207 00:24:41,150 --> 00:24:46,660 of game changing recovery trial that was also based on that on the small molecule front, 208 00:24:46,670 --> 00:24:54,740 that was actually the direct acting antiviral from that was not very much that was able to be repurposed. 209 00:24:55,620 --> 00:24:59,450 And what does the protease actually do and how does that help the virus things? 210 00:24:59,840 --> 00:25:12,380 Okay, so so when the virus enters the cell, it enters through the spike protein and over to receptors ace2 and ramps to it gets kind of into the cell. 211 00:25:12,860 --> 00:25:19,160 And as soon as it gets into the cell, it unpacks itself and comes into the cytoplasm. 212 00:25:19,550 --> 00:25:24,770 And once it sits into the cytoplasm, it gets translated into not to long party proteins. 213 00:25:25,790 --> 00:25:32,399 They are called open reading from one A and one B, and with this you can't really do very much. 214 00:25:32,400 --> 00:25:39,469 But all like all the proteins that the virus needs to replicate and make more virus or contain them, 215 00:25:39,470 --> 00:25:52,010 these kind of two very long bits of material and the protease kind of folds itself up and starts keeping itself out of this open reading frame, 216 00:25:52,220 --> 00:25:55,010 but it also releases all the other proteins. 217 00:25:55,010 --> 00:26:01,940 The main protease protease cleaves the party protein in 11 different state, and they make them different places. 218 00:26:02,180 --> 00:26:09,140 And then there's another protease which is called the paper like protease, and that feeds in three different places. 219 00:26:09,500 --> 00:26:18,620 And the main protease has some similarity to other viral proteases that we knew from experience can be targeted. 220 00:26:19,010 --> 00:26:30,629 So we know from experience that HIV proteases have been targeted and this has an effect on viral load and HCV proteases have been targeted. 221 00:26:30,630 --> 00:26:34,880 So hepatitis C virus parties have been targeted and that has no sex involvement. 222 00:26:35,270 --> 00:26:40,580 So it was one of the obvious targets to go after. The other one with clinical validation. 223 00:26:40,610 --> 00:26:45,920 So where we knew that if we inhibit that will have an impact on viral load is the polymerase. 224 00:26:46,280 --> 00:26:49,400 It was those were the kind of two proteins where we kind of knew. 225 00:26:50,910 --> 00:26:54,440 We have a chance. Basically, if we target those, we'll have a chance. 226 00:26:54,450 --> 00:27:00,059 And one of the other obvious things was that these two proteins were taken and a lot of screens 227 00:27:00,060 --> 00:27:06,299 were running against them to see whether some of the existing inhibitors find and inhibit things. 228 00:27:06,300 --> 00:27:07,410 And unfortunately. 229 00:27:08,720 --> 00:27:17,840 And we didn't really have any kind of hepatitis C protease inhibitors, for example, that work very well against the proteins and the polymerase. 230 00:27:17,840 --> 00:27:19,580 That was a different game. There were. 231 00:27:19,970 --> 00:27:29,400 And so Remdesivir, for example, was developed for and I think over there originally and it wasn't on the market yet. 232 00:27:29,410 --> 00:27:34,430 It wasn't. But the whole development process had already been done. 233 00:27:34,820 --> 00:27:42,140 And similarly, from olaparib, which was developed for RNA viruses that had been developed originally, I think it was for. 234 00:27:43,510 --> 00:27:49,660 Venezuelan equine encephalitis was so easy. 235 00:27:51,250 --> 00:27:55,900 And so a lot of the kind of development for that compound was also done. 236 00:27:56,140 --> 00:28:04,000 And actually, the inhibitors that we have now Paxos it and which was developed by Pfizer and that 237 00:28:04,000 --> 00:28:10,570 piggybacked on quite old drug discovery efforts that were discontinued previously. 238 00:28:10,990 --> 00:28:21,760 So they used data that they had from source one and they use data and that effort even used data based on rhinoviruses. 239 00:28:22,060 --> 00:28:32,400 So there's kind of quite a lot of background research on the proteases that kind of show that if you target this and that you can do it efficiently, 240 00:28:32,410 --> 00:28:34,720 you may have a good chance to see something. The. 241 00:28:35,980 --> 00:28:44,560 However, we didn't know until the phase two Phase two, three data from Pfizer actually came out that this was a target that was ten or 12 inches. 242 00:28:45,040 --> 00:28:51,790 So it took another one and a half years to actually say, yes, this is a worthwhile target, to be honest. 243 00:28:52,270 --> 00:28:58,030 Prior to that, we just were able to say it has worked on other viruses with some of the proteases, 244 00:28:58,330 --> 00:29:03,489 but we weren't sure that it was going to work, but it wasn't a diversion. 245 00:29:03,490 --> 00:29:07,860 So you would think that you'd be telling me how the Chinese have approached Frank analytically? 246 00:29:08,950 --> 00:29:12,370 Yeah, yeah. I kind of ran away from that. 247 00:29:12,820 --> 00:29:23,320 So the Chinese had contacted Diamond Lifestyles and specifically Frank's group and said, Look, we're going we've got the construct that was protease. 248 00:29:23,320 --> 00:29:29,020 So the material that you need to run this this experiment, which is called a Fresnel screen. 249 00:29:29,680 --> 00:29:39,550 And the aim of this experiment is to basically look at the protein and 3D shape and find the active site. 250 00:29:39,580 --> 00:29:51,490 So the site where the protease cleaving and then you use small bits of drug, which are called fragments, to basically paint that active site. 251 00:29:51,850 --> 00:29:55,840 So what you want to know is how would an inhibitor look like? 252 00:29:56,500 --> 00:30:00,850 And that would fit very snugly into this active site. 253 00:30:02,100 --> 00:30:08,370 And if you use existing drugs to do that, you've got a good chance that one part of the drug might make it, 254 00:30:08,370 --> 00:30:14,490 but the other part hangs out on the other side. Or maybe it sits in a kind of weird position. 255 00:30:14,850 --> 00:30:18,480 But if you use small particles of a compound, 256 00:30:18,810 --> 00:30:28,560 then you have a better chance of mapping out the different positions very nicely and finding something that will bind with a very high affinity. 257 00:30:28,890 --> 00:30:32,580 And also kind of you want to kind of find the basically. 258 00:30:32,940 --> 00:30:38,310 And by using lots of small bits of drug and putting them into the crystal structure, 259 00:30:39,120 --> 00:30:48,480 you may be able to kind of get a better idea of how this kind of active binding site is setup and what you may need to do to map it out properly. 260 00:30:49,050 --> 00:30:52,830 So this experiment is called a fragment screen, 261 00:30:53,220 --> 00:31:00,810 and that is exactly the kind of experiment that they run at the time like this for a whole lot of different human proteins, but also viral proteins. 262 00:31:01,440 --> 00:31:10,290 And however, to run this experiment, you kind of need the protein construct that you then make your protein from, 263 00:31:11,220 --> 00:31:19,950 and you would follow the protein the right way and have a crystal form to then be able to run this experiment under the x ray. 264 00:31:19,970 --> 00:31:27,000 You. And the experiments with just the crystal have been run in China and the Chinese. 265 00:31:27,000 --> 00:31:28,860 We contacted them and said, Look, 266 00:31:28,860 --> 00:31:38,939 we can ship you all the constructs over and we can even ship you the protein over for you to run the fragments cream and the Cooper Diamond said, 267 00:31:38,940 --> 00:31:41,940 Yes, we will drop everything and ship us the protein. 268 00:31:43,110 --> 00:31:50,310 However, the Chinese group was hit by the local lockdown and they weren't able to actually ship anything because everything was just shut. 269 00:31:50,820 --> 00:31:54,630 And so they just gave the diamond group the information. 270 00:31:54,660 --> 00:32:00,240 And Martin Walters group of diamond expressed the protein very, very fast. 271 00:32:00,690 --> 00:32:08,060 And the team at Diamond literally dropped everything and rammed this fragment screen as the absolute priority. 272 00:32:08,430 --> 00:32:15,600 They generate a lot of different structures, and we're actually really lucky that the fragment screen ran so well. 273 00:32:16,800 --> 00:32:22,320 And they've got a lot of different fragments and decided that this information has to go out 274 00:32:22,320 --> 00:32:27,720 in the open as fast as possible and to enable drug discovery efforts to use that information. 275 00:32:28,500 --> 00:32:31,640 And. At that point. 276 00:32:31,790 --> 00:32:40,400 That was where the real difficulty started because the Diamond team had run this kind of fragment screen at very high throughput. 277 00:32:41,030 --> 00:32:49,100 But what is not yet established is technology to get from all these different fragments to an active compound. 278 00:32:49,490 --> 00:32:52,490 So the so called interlinked finding. 279 00:32:52,820 --> 00:32:58,100 So you've got different hints. So you've got some idea of what might fit into that finding size. 280 00:32:58,820 --> 00:33:02,450 But none of these hits independently will actually inhibit the protein. 281 00:33:02,720 --> 00:33:12,350 We'll have to make sure that this kind of thing gets going together or joined up and make a compound that actually inhibits the protein, 282 00:33:12,440 --> 00:33:18,230 survives of animal or human and and actually has an effect on the compound. 283 00:33:18,240 --> 00:33:25,370 And that usually is a very, very long process of 5 to 10 years to kind of get to that point. 284 00:33:27,550 --> 00:33:36,700 And what? And Frank and a couple of his collaborators that have been trying to kind of work on techniques to 285 00:33:36,700 --> 00:33:42,370 address this particular problem previously have decided at that point is we actually are out of orbit. 286 00:33:43,090 --> 00:33:46,510 We don't really know how to get from a segment to a compound, 287 00:33:47,230 --> 00:33:55,330 but we can outsource or crowdsource this problem and basically ask a lot of medicinal chemists to contribute to this. 288 00:33:55,750 --> 00:34:04,210 And the kind of key problem is that the moonshot at that point decided, okay, we're crowdsourcing this, we're doing something completely new. 289 00:34:04,660 --> 00:34:14,440 Everything is going to be in the open. We're going to ask medicinal chemists all over the world to look at this information and tell us their ideas. 290 00:34:14,440 --> 00:34:21,010 And what we will do is we will make sure that these compounds are actually ranked and made and tested. 291 00:34:21,340 --> 00:34:29,770 So near London and at the Vital Institute and Israel said, if you give me the protein I can set up, and I say, they've got a high acid facility. 292 00:34:30,070 --> 00:34:33,690 And Chris Schofield here in Oxford at the Chemistry Institute said the same. 293 00:34:33,700 --> 00:34:40,780 They set up a different kind of breathalyser with two different acids, one then all funny, the posterior, 294 00:34:41,200 --> 00:34:50,680 which is a newly found biotech company and said, look, guys, we can't run this all on Google sheets or Google Docs. 295 00:34:51,040 --> 00:34:57,700 We'll need to kind of make a better plan. So within one and a half days, one of his programmers, Matt, 296 00:34:58,000 --> 00:35:06,790 have set up a website that all the medicinal chemists could submit their ideas to and then. 297 00:35:08,130 --> 00:35:16,260 Very early in the process. And medicinal chemists that work for AstraZeneca antibiotics previously called Ed Griffin, 298 00:35:16,410 --> 00:35:23,160 now works for a small biotech as well and said, I can coordinate this efforts and kind of try and help to rank the ideas. 299 00:35:23,850 --> 00:35:36,570 He came involved and then John Cordero, who's working as in this case D.C. and New York and he has a I technique or I'm actually. 300 00:35:37,640 --> 00:35:44,060 He's running a machine learning modelling that predicts which compounds will fit very 301 00:35:44,060 --> 00:35:48,830 well into the active binding site and have a high likelihood of being a good inhibitor. 302 00:35:49,340 --> 00:35:55,130 And he uses a technique called FFP for these calculations, he says. 303 00:35:55,430 --> 00:35:58,820 I'm going to do exactly that for the Moonshot. 304 00:35:58,850 --> 00:36:06,980 These are super complicated mathematical modelling calculations that I don't completely understand. 305 00:36:07,400 --> 00:36:11,630 But what I do know is that he built the biggest, the world's biggest supercomputer, 306 00:36:12,920 --> 00:36:18,530 running these calculations for the moonshot and through a consortium that is called folding a 307 00:36:18,530 --> 00:36:24,860 term which basically uses GPUs from home computers worldwide to kind of group them all together. 308 00:36:24,890 --> 00:36:28,970 So that was that was kind of the best thing when they came up with these ideas and then said, 309 00:36:29,540 --> 00:36:37,250 we're going to tweet about this and then we're going to find some money between Cycle Christian's to try to get these companies made. 310 00:36:37,880 --> 00:36:41,290 And so they all kind of went to that research budget. 311 00:36:41,300 --> 00:36:48,800 And so what they could redistribute and to making sexual compounds a little bit of money here and there. 312 00:36:48,800 --> 00:36:53,150 And then the kind of crucial step was the Ukrainian company. 313 00:36:54,720 --> 00:37:00,090 Joining the picture and saying we can actually synthesise these compounds for you at cost. 314 00:37:00,960 --> 00:37:08,340 So they were the kind of crucial contributor, not only because they were actually making the compounds, 315 00:37:08,340 --> 00:37:12,450 you could have outsourced to other compounds, but they've also picked up all the logistics. 316 00:37:12,840 --> 00:37:17,280 So they sent out the compounds as soon as they were made to Diamond Light source 317 00:37:17,280 --> 00:37:22,440 to run the code crystal structures and also coordination around the project. 318 00:37:22,920 --> 00:37:28,360 And that was kind of the birth of the neutron. I think that was the eight months, 2020 where where this all happened. 319 00:37:28,380 --> 00:37:33,270 And then a lot of the response to to all these tweets was overwhelming. 320 00:37:33,270 --> 00:37:40,980 So we submitted to the website, and I think I was going to I was going to sort of bring that up because this I mean, 321 00:37:41,220 --> 00:37:45,030 the scientific community does use Twitter quite a lot now. Can you hear me? 322 00:37:45,360 --> 00:37:46,860 Yes, I've just started using a microphone. 323 00:37:49,380 --> 00:37:55,770 But the this was really off the scale, wasn't it, the extent to which Twitter expanded the moonshot project? 324 00:37:56,430 --> 00:38:03,270 Yes, it was. It was definitely a much bigger response than the original team had expected. 325 00:38:03,720 --> 00:38:05,870 There were suddenly like that. 326 00:38:06,000 --> 00:38:14,879 I remember at that time, and I only know that because I was actually doing the homeschooling, feeling really unwell with COVID at home at the time. 327 00:38:14,880 --> 00:38:24,230 But basically Frank was on 3 to 4 logistical conversations every day trying to set up this operation. 328 00:38:24,240 --> 00:38:31,889 They were overwhelmed with responses at the start and and trying to kind of get this effort 329 00:38:31,890 --> 00:38:37,470 off the ground was there was something that they had that none of them had ever done before. 330 00:38:38,160 --> 00:38:44,069 So it took quite a lot of coordination between all these different sites because 331 00:38:44,070 --> 00:38:49,860 everyone was sitting at home and all these medicinal chemists that saw these tweets, 332 00:38:50,280 --> 00:38:54,780 looked at the data and said, oh, yeah, I actually I can imagine doing this. 333 00:38:54,870 --> 00:39:00,030 This would do something. And so, yeah, the response was overwhelming. 334 00:39:00,390 --> 00:39:04,200 And, and the buy in from the scientific community was overwhelming. 335 00:39:04,200 --> 00:39:10,050 I think it was a unique situation because this came exactly the week where everyone was locked down suddenly. 336 00:39:10,590 --> 00:39:15,600 And so this is so we're still only in the end of March 2020. 337 00:39:15,900 --> 00:39:19,860 Yes. Yes. This was so the tweet went out on the 18th of March. 338 00:39:19,860 --> 00:39:26,640 And I think the lockdown on the 25th or 23rd, 23rd was yes, it was exactly that time. 339 00:39:27,120 --> 00:39:32,459 And I'm actually uniquely because so this is this is where the kind of uniqueness 340 00:39:32,460 --> 00:39:38,340 of the moonshot every single compound was online and visible at conception. 341 00:39:39,460 --> 00:39:43,010 So everything was completely in the open. 342 00:39:43,030 --> 00:39:49,330 Everyone could see the compound. And as soon as we had an end, as soon as it was made and we had an enzyme result available, 343 00:39:49,810 --> 00:39:58,570 it was on the website and visible to everyone so anyone could trace the information and see what had happened, 344 00:39:58,960 --> 00:40:06,000 whether their compound was being round to be made because we couldn't make all the submissions. 345 00:40:06,010 --> 00:40:13,360 We ran the submissions first and then made the top ones that we both work, both because of budget constraints, 346 00:40:13,360 --> 00:40:23,020 but also because we didn't have enough synthesis capacity even at a company like me to make actually synthesising those compounds. 347 00:40:23,790 --> 00:40:27,940 And the other thing that happened is that. 348 00:40:29,140 --> 00:40:40,570 A lot of employees of pharmaceutical companies that would usually not be able to contribute to anything outside the company was uniquely able to 349 00:40:40,570 --> 00:40:48,850 contribute to this because there wouldn't be any patents on this chemical matter because everything was open and disclosed right at the start. 350 00:40:49,660 --> 00:40:53,710 So that was something that was. 351 00:40:54,620 --> 00:41:05,780 Unprecedented, I think. And the other thing that was unprecedented was the interest from pharmaceutical companies to contribute in-kind. 352 00:41:05,960 --> 00:41:16,070 So for example, we had a whole team of medicinal chemists that came to the weekly meetings and were able from the company 353 00:41:16,070 --> 00:41:24,440 to contribute a slice of their time to synthesise or to design compounds for the moonshot effort. 354 00:41:24,770 --> 00:41:29,740 And that's in addition to the people that were sitting at home and just doing this for fun and. 355 00:41:30,940 --> 00:41:34,929 And I think this the timing was was unique. 356 00:41:34,930 --> 00:41:41,379 And also the way of doing it was unique and was very much kind of spurred by the pandemic because this 357 00:41:41,380 --> 00:41:49,270 was something of interest to everyone and which which probably won't be repeated in this way ever again. 358 00:41:49,510 --> 00:41:57,790 Or maybe it might be. Let's hope not, because let's hope we won't have the situation again where we'll all have to lockdown because of the pandemic. 359 00:41:58,810 --> 00:42:04,120 So what you've described is like a kind of massive filter that's taking in an enormous amount of material at the top. 360 00:42:04,480 --> 00:42:14,750 And what you want to get out at the bottom is a drug that can be used to both to prevent and to treat the virus. 361 00:42:14,770 --> 00:42:17,440 Is that the idea? And sorry, another question at this point. 362 00:42:17,440 --> 00:42:23,379 This is you mentioned that while all this was going on, you were homeschooling at home, but you were very much part of the moonshot project. 363 00:42:23,380 --> 00:42:35,020 So what was your role? So I joined a little later and I wasn't involved at the right at the start, and I was actually at that time helping Nicole's. 364 00:42:35,020 --> 00:42:42,670 It's my news of our just to get Oxford to set up a screening facility, and that was for repurposing compounds. 365 00:42:43,090 --> 00:42:53,229 So I have a slightly different role and basically all my normal work which was on epigenetic inhibitors shut 366 00:42:53,230 --> 00:43:00,430 down completely because then all all non-essential corporate work was deprioritized and didn't happen. 367 00:43:01,120 --> 00:43:07,750 But I had worked on viruses before quite a lot and actually had grown viruses as well. 368 00:43:07,780 --> 00:43:16,000 So I had the necessary expertise to set up help set up that part of the kind screening program. 369 00:43:16,300 --> 00:43:19,530 And right at the start, the mutual obviously didn't have any compounds. 370 00:43:19,540 --> 00:43:24,490 It takes time to make these compounds. So I wasn't involved with that part of the work. 371 00:43:24,730 --> 00:43:30,280 I was trying to set up a screening facility with with Nicole and William James. 372 00:43:31,030 --> 00:43:37,570 Obviously, that also didn't happen that fast because we had to kind of get some virus first, make sure the virus is growing. 373 00:43:37,840 --> 00:43:41,950 And my my role in this kind of whole setup was very organisational. 374 00:43:41,950 --> 00:43:55,810 So I looked through all the nominations for screening, ranked those for the process and, and then and ultimately able to send you the screening. 375 00:43:56,080 --> 00:44:01,030 So it was a very, very different role. But as soon as mutual funds were available, 376 00:44:01,300 --> 00:44:07,540 I was then able to help to coordinate all the cellular screening of the capacity that we had at Oxford was insufficient in any way. 377 00:44:07,900 --> 00:44:16,140 So we felt like we had a massive buy in from anti-viral groups worldwide that said, yes, we can screen some compounds for you. 378 00:44:16,150 --> 00:44:21,610 Like some of them were fairly high throughput because of the technologies they use them because of the access they use. 379 00:44:21,940 --> 00:44:28,510 And some of them were very low throughput, so some of them screened maybe five or ten compounds, some of them screened hundreds. 380 00:44:28,960 --> 00:44:33,730 And so it was very dependent on what was exactly used in that. 381 00:44:34,870 --> 00:44:36,370 And that particular set up. 382 00:44:36,730 --> 00:44:45,710 But then my role kind of evolved, and most of the work that I was doing, I was doing at night because during the day I was homeschooling. 383 00:44:48,060 --> 00:44:54,180 And it wasn't directly linked to the moon shot. But at the time that the moon shot actually had some of the active compounds, 384 00:44:54,480 --> 00:45:00,690 we had all setup running to contribute to the neutrons that kind of all came together. 385 00:45:00,690 --> 00:45:07,770 So right at the start when everything was set up, I had a wolf, a commenting role, 386 00:45:09,990 --> 00:45:14,820 and I was just very interested to see how everything was working out. 387 00:45:15,210 --> 00:45:22,230 But my actual daytime job was very different and only and I would say probably early June. 388 00:45:23,510 --> 00:45:29,450 That was when I when I started contributing to the job, because that was when they had active companies coming through. 389 00:45:31,520 --> 00:45:38,510 Does that answer your question? Yes, it does. Yes. Yes. And and so how did it develop from from there? 390 00:45:39,050 --> 00:45:44,420 How how what what what what have been the main outcomes of the project? 391 00:45:44,920 --> 00:45:48,880 Okay. So this is now June 20, 21st. 392 00:45:49,250 --> 00:45:56,240 It became clear and during that time that we were on to something. 393 00:45:56,390 --> 00:46:03,490 So we at that point had several medicinal cannabis, hemp seeds, several series, 394 00:46:03,500 --> 00:46:08,090 so different compounds that looked different that showed cellular activity. 395 00:46:08,600 --> 00:46:13,490 And at that time, we had to decide which of those to take forward. 396 00:46:14,030 --> 00:46:20,059 So we were trying to kind of screen as widely as possible across the series to make sure that we 397 00:46:20,060 --> 00:46:26,150 had and that we chose the right one to take forward because of the kind of financial constraints, 398 00:46:26,480 --> 00:46:30,020 we basically had to get our compounds a lot more potent. 399 00:46:30,050 --> 00:46:31,879 At that point, they were potent, 400 00:46:31,880 --> 00:46:39,950 but they weren't at a point where we would expect them to have a good antiviral effect once they are going to be put into human. 401 00:46:40,970 --> 00:46:49,190 So we had to drive the potency down and at the same time, we also had to kind of figure out whether these were the kind of compounds that would. 402 00:46:51,550 --> 00:46:54,300 Survive and and a human. 403 00:46:54,700 --> 00:47:03,010 And for that, you do a range of tests and these tests are usually done at contract research organisations and they're very expensive. 404 00:47:03,730 --> 00:47:10,030 So we had to start looking for money to pay for that simple grants. 405 00:47:10,690 --> 00:47:23,020 And up to January 2021 and it was actually extremely difficult to come to find money to fund a novel drug discovery because we were still hoping. 406 00:47:24,090 --> 00:47:29,100 To solve the problem completely with vaccines. And actually, we've done tremendously. 407 00:47:29,100 --> 00:47:34,570 We're so lucky that we have done so tremendously well with the vaccines and. 408 00:47:35,730 --> 00:47:42,000 And we were also still hoping to find some kind of repurposed drugs that so some kind 409 00:47:42,000 --> 00:47:46,620 of existing drug that would get us around the fact of having to develop a new drug. 410 00:47:47,280 --> 00:47:52,470 And it was clear at the time that there were a few companies working on this. 411 00:47:52,920 --> 00:48:00,960 But what we didn't know was whether any of these new compounds would actually work and how quickly they would be available. 412 00:48:01,620 --> 00:48:06,419 And the effort that Pfizer has put down is actually absolutely unprecedented. 413 00:48:06,420 --> 00:48:12,250 And they invested a lot of money into it upfront with a very high risk to get to that. 414 00:48:12,270 --> 00:48:20,139 So we were never in the position to match that. But. What wasn't clear is what would be coming afterwards. 415 00:48:20,140 --> 00:48:25,640 And the other experience that we do have from anti-viral drugs is there will be resistance developing. 416 00:48:25,660 --> 00:48:29,290 So we will always have new viral strains coming. 417 00:48:29,290 --> 00:48:39,760 We'll always have to have not only one drug available, but several drugs available to be able to combat a new circulating strain. 418 00:48:40,150 --> 00:48:44,470 And we can't predict very well what that is. That new strain is going to be. 419 00:48:44,740 --> 00:48:46,360 We may have some idea, 420 00:48:46,360 --> 00:48:54,430 but basically the best that we can do is have an arsenal available that targets different proteins and targets them in different ways. 421 00:48:54,670 --> 00:49:02,440 So that's we knew that at that point, but we weren't actually able to move very far ahead without additional funding. 422 00:49:03,540 --> 00:49:10,319 And so my main role in the setback in the coming months was actually trying to 423 00:49:10,320 --> 00:49:15,690 look for funding and trying to write grants that would fund the next steps, 424 00:49:16,170 --> 00:49:18,870 but also to ask for in-kind contributions. 425 00:49:18,870 --> 00:49:26,699 And at that point, we were actually much more successful asking for in-kind contributions from other pharmaceutical companies, 426 00:49:26,700 --> 00:49:36,450 then asking for funding. So a lot of our funding approaches were deemed as not in scope because up to 427 00:49:36,450 --> 00:49:42,719 2021 novel drug discovery was not in scope for a lot of the grant applications. 428 00:49:42,720 --> 00:49:47,570 It was only vaccine development, clinical trials and repurposed drugs. 429 00:49:48,450 --> 00:49:55,380 That only changed in 2021, and that's when we started talking to the Wellcome that is now funding us. 430 00:49:55,800 --> 00:50:00,390 And for this effort they started funding as of July 2021. 431 00:50:01,360 --> 00:50:04,390 And for the making, the optimisation and preclinical stages. 432 00:50:04,870 --> 00:50:11,949 What happened prior to that is that several pharmaceutical companies bought into the project and often alongside 433 00:50:11,950 --> 00:50:18,970 internal efforts that they are pushing themselves and provided us with huge amounts of in-kind contribution. 434 00:50:19,890 --> 00:50:23,910 And in all cases and contributions were mainly. 435 00:50:26,740 --> 00:50:34,990 And Acme testing. And ACME testing is something that everyone who works on drug discovery bodes very well what it is. 436 00:50:34,990 --> 00:50:38,500 But it's a it's a pretty theoretical concept for everyone else. 437 00:50:38,860 --> 00:50:42,160 So basically, it's a range of lots of different tests. 438 00:50:42,550 --> 00:50:48,370 How do you how do you spell it? It's an acronym, say BMB stands for absorption. 439 00:50:48,520 --> 00:50:52,569 So how does the drug get into the body and distribution? 440 00:50:52,570 --> 00:51:01,330 How does the drug behave when it gets into the body, or does it tend to go into kind of the fat tissue? 441 00:51:01,330 --> 00:51:05,920 Does it tend to kind of stay in the blood volume then metabolism? 442 00:51:06,130 --> 00:51:13,540 So basically what happens when the drug is the liver and which enzymes if the drug, how quickly does that happen? 443 00:51:13,960 --> 00:51:17,830 Does it block enzymes and then elimination? 444 00:51:17,830 --> 00:51:21,280 So basically, how quickly does it leave the body again? 445 00:51:21,550 --> 00:51:24,640 And which transporters does it used to leave the body again? 446 00:51:24,730 --> 00:51:33,590 So at a end. And all like there's one way to test it, which is to give it to someone and see what it does. 447 00:51:34,370 --> 00:51:40,850 Obviously, you can't really do that with a new drugs because it may be toxic and they don't want to give it to someone at this point. 448 00:51:41,450 --> 00:51:49,310 But what you can do is run a lot of kind of tests to see what each of these parameters may look like. 449 00:51:49,550 --> 00:51:55,010 By the time you've actually put this into a human being and for example, 450 00:51:55,190 --> 00:52:03,350 you may use some liver cells and put the compound onto the liver cells and see how long it survives when it gets chopped up straightaway. 451 00:52:04,010 --> 00:52:10,180 Or you may add it to some transporters and see how quickly it gets transported in and out of the cell. 452 00:52:11,030 --> 00:52:14,810 Or you may look at plasma protein binding. 453 00:52:15,020 --> 00:52:25,130 So you will have some human serum and add the compound to it and then see how much of the compound gets bound to the serum and how much things free. 454 00:52:25,970 --> 00:52:31,129 And all these are very important parameters to try to work out how whether you compound might be 455 00:52:31,130 --> 00:52:36,470 something that may behave very well in humans or might be something that is actually complete and. 456 00:52:37,190 --> 00:52:40,010 You can give it to a but it may not even be taken out. 457 00:52:40,650 --> 00:52:49,340 And so this is and all these tests are very, very important, the correct discovery, but at the same time, very, very expensive to run. 458 00:52:49,850 --> 00:52:57,950 And so we were very lucky because Novartis, the company basically as in-kind contribution, 459 00:52:57,950 --> 00:53:08,950 gave a whole block set of the tests to the COVID moonshot as an important contribution and enabled us to run this these tests of that company. 460 00:53:09,340 --> 00:53:14,300 Roughly how many compounds we are looking at at this stage. Hundreds. 461 00:53:14,620 --> 00:53:25,099 Hundreds. I'm sad to say. The thing is, it's also it's not not that we make all the compounds upfront and say we are going to test movies. 462 00:53:25,100 --> 00:53:32,060 What, it's a cycle. So you make the compound, you see, oh, this works really well, is really potent. 463 00:53:32,540 --> 00:53:39,139 That actually it's cleared too quickly and it doesn't really behave that well 464 00:53:39,140 --> 00:53:45,170 when you actually exposed to the liver and then the medicine covers the clever. 465 00:53:46,040 --> 00:53:50,480 So if we change this little bit here on the compound, that may help us. 466 00:53:50,720 --> 00:53:54,530 So you make that called politics to the game. I see it's not token. 467 00:53:54,570 --> 00:53:58,700 So we have to try loads, different ones to get the potent one again. 468 00:53:58,700 --> 00:54:07,460 And then you test the game and say, okay, so this time we came up with something that behaves well in the liver and is potent, 469 00:54:08,030 --> 00:54:13,769 but actually it there may be something else wrong with this. 470 00:54:13,770 --> 00:54:16,850 So and actually if not taken up into the body. 471 00:54:18,060 --> 00:54:24,660 So it's it's an iterative cycle and you kind of narrow down and test different things. 472 00:54:25,500 --> 00:54:33,030 And at the end, we tested probably around 600 or 700 compounds in different tests of different combinations. 473 00:54:33,570 --> 00:54:38,430 Overall, we generated 2300. 474 00:54:38,700 --> 00:54:45,149 We actually made 2300 compounds to get to the kind of three compounds that we are now evaluating 475 00:54:45,150 --> 00:54:51,000 closer and that we still we're still not at the point where we can say this is our final compound, 476 00:54:51,000 --> 00:54:56,850 but we can take it to. So if you said three, you've gone from hundreds to three. 477 00:54:56,850 --> 00:55:04,559 Is that what you said at the three stage? And actually, this is what the Wellcome Trust is funding us to do, 478 00:55:04,560 --> 00:55:13,110 is get from 3 to 1 compound and actually have have one compound available that we are planning to put into clinical trials next year. 479 00:55:13,350 --> 00:55:19,379 So that's where we are currently at. We are running all the preclinical scale up, 480 00:55:19,380 --> 00:55:27,150 which basically means making a whole lot of other compounds so that you can like at the end of the month to actually give it to patients. 481 00:55:27,660 --> 00:55:36,510 And a whole lot of process chemistry is required for that because what you have to imagine is we made tiny, 482 00:55:36,510 --> 00:55:43,470 tiny amounts for the use of this compound to look at cellular assays and all these tests. 483 00:55:44,010 --> 00:55:47,580 But if you were to give it to a patient, we actually do quite a lot of it. 484 00:55:49,480 --> 00:55:52,660 How much exactly is what we're trying to work out at the moment? 485 00:55:53,710 --> 00:55:58,630 And this is this is called process chemistry. So that's that's part of what's happening right now. 486 00:55:58,960 --> 00:56:05,590 And there's a whole lot of other process chemistry and you'll hear a lot more of that with the vaccine people is what 487 00:56:05,590 --> 00:56:13,840 comes after your phase one to actually provide hundreds of thousands or ten of 400,000 people with these compounds. 488 00:56:13,840 --> 00:56:19,270 So that's a whole different ballgame again. And that's where pharmaceutical industry is actually pretty good at. 489 00:56:20,740 --> 00:56:25,810 So so do you have do you have a pharmaceutical industry partner who's taking all that on? 490 00:56:25,840 --> 00:56:32,290 No. No, because we haven't got the patent. So it will be very difficult to make money from our. 491 00:56:33,040 --> 00:56:36,370 We'll be more like a generic company. 492 00:56:36,520 --> 00:56:40,720 Yeah. That would be interested in this kind of product. 493 00:56:42,340 --> 00:56:48,190 So that's, that's very much what is happening in our strategic discussions at the moment with the eyes. 494 00:56:48,550 --> 00:56:56,200 How can we come up with a set up that entices a generic partner to work with us on this? 495 00:56:56,680 --> 00:57:03,969 Because it is quite unlikely that a pharmaceutical company will have enough interest in a compound 496 00:57:03,970 --> 00:57:09,430 that doesn't come with the patents where you can't earn a lot of money off the compound. 497 00:57:09,790 --> 00:57:18,070 So basically this this would be a cheap compounded generic from the start that is available to the masses. 498 00:57:18,280 --> 00:57:23,410 So that's I mean, in a way, it's a unique selling point in terms of drug access. 499 00:57:23,830 --> 00:57:27,760 But if you want to add a lot of money up a compound, it's not particularly attractive. 500 00:57:28,910 --> 00:57:32,780 Mm hmm. Very interesting. So I've got so two questions of that. 501 00:57:33,120 --> 00:57:38,900 So one is, first of all, what how do you envisage this company, if it works, if it passes all these tests? 502 00:57:39,290 --> 00:57:43,210 How do you envisage it being used in the future? 503 00:57:43,220 --> 00:57:51,530 Is it is it would you give it prophylactically or is it something you would give to people who have symptoms or how would it actually be used? 504 00:57:51,800 --> 00:57:55,700 So if you think about like natural infection of courage, 505 00:57:56,180 --> 00:58:02,660 what happens at the start is you get exposed to the virus and then you get a viral load increasing. 506 00:58:02,930 --> 00:58:10,700 And then right that what happened right at the start is because we didn't have any natural immunity, we didn't have any vaccine responses, 507 00:58:11,090 --> 00:58:22,670 and we did see the kind of innate, innate immune response kind of rising up and generating this kind of very acute picture. 508 00:58:23,540 --> 00:58:32,000 And that is where a lot of the immunosuppressants worked kind of when you got into hospital and they keep dampening down that immune response. 509 00:58:32,450 --> 00:58:37,190 What we are talking about is this kind of first increase in viral load. 510 00:58:37,550 --> 00:58:41,960 So this is a drug that would have to be given very early after infection. 511 00:58:42,900 --> 00:58:46,420 And it's similar to Paxton's actually exactly the same sex marriage. 512 00:58:46,770 --> 00:58:49,320 You have to give it as early as possible. 513 00:58:50,470 --> 00:59:01,240 It works best if you give it within the first five days and it will decrease your likelihood of developing a innate immune response towards the virus. 514 00:59:01,510 --> 00:59:11,049 And that then brings you into hospital or if you die at the what is happening at the moment is that this picture is changing completely 515 00:59:11,050 --> 00:59:19,180 anyway because we've got a lot of natural immunity that is developing with all the different variants that's coming through the population. 516 00:59:20,500 --> 00:59:26,500 And a lot of people are not dying anymore, but still getting quite sick from the viral load. 517 00:59:26,860 --> 00:59:34,830 So what we would envisage is to use this similarly to how we use a lot of flu or Tamiflu. 518 00:59:35,920 --> 00:59:41,290 So right at the start, when, you know you have been exposed, you may have just had a positive test. 519 00:59:41,590 --> 00:59:47,110 That's when you take the drug. And you would hope that that would ameliorate your symptoms. 520 00:59:47,380 --> 00:59:54,910 But also, if you are at risk, patients that may not have an immune response yet was not eligible to the vaccine at that point. 521 00:59:54,910 --> 01:00:04,080 Would. Lower your risk of hitting a hospitalisation point would lower your risk of death. 522 01:00:04,380 --> 01:00:10,080 And in the kind of worst case scenario. Here again, 523 01:00:10,110 --> 01:00:20,700 we're kind of moving into a bit of a difficult space because what we and the clinical trials and points that we are looking at currently still have 524 01:00:21,510 --> 01:00:33,389 death and mortality and clinical trials are hospital admissions as the primary end point or even just a drop in the oxygen load or clots developing. 525 01:00:33,390 --> 01:00:42,750 So everything that kind of comes in the later stages. We may have to look very different end points in one or two years time to come where we look at 526 01:00:42,750 --> 01:00:49,650 symptom duration or just lowering of viral load as a kind of public health intervention in a way. 527 01:00:50,040 --> 01:00:54,210 And so this is this is a very rapidly changing field. 528 01:00:54,600 --> 01:00:59,579 And the part of that that we can't predict is what is going to happen if there's 529 01:00:59,580 --> 01:01:03,340 a virus coming along that evades immunity much more than public health, 530 01:01:03,390 --> 01:01:13,950 then, for example, we still expect the protease inhibitors to work quite well across the different coronavirus strains. 531 01:01:14,250 --> 01:01:23,399 So that's something that we know from. And part of genetic sequencing is that the protease is relatively constant. 532 01:01:23,400 --> 01:01:28,890 It's not quite as close to the race, but it's more conserved than, for example, spike protein. 533 01:01:29,250 --> 01:01:38,820 So it's a good likelihood that the protease inhibitors will play a role for new strains developing and that escape the immune response. 534 01:01:39,780 --> 01:01:46,770 But it's a very changing and unclear picture, so it's a good question. 535 01:01:48,190 --> 01:01:52,750 My ensemble will look probably very different to what my object would look like in three months. 536 01:01:54,910 --> 01:01:58,540 But you have at least you have the funding to take it forward. So that's really good. 537 01:01:58,890 --> 01:02:06,700 So looking at looking at this that the the the amazing process that you've all gone through, 538 01:02:07,390 --> 01:02:12,250 do you see this as a model for future drug discovery or do you think this was a one off? 539 01:02:15,010 --> 01:02:19,089 I think that's how it went and the speed it went. 540 01:02:19,090 --> 01:02:24,430 That was probably a one off. However, I think we can learn. 541 01:02:24,760 --> 01:02:36,070 We can take very important bits and pieces from it and apply them to different diseases and also to similar diseases, but for different compounds. 542 01:02:36,250 --> 01:02:42,280 And part of that is actually what we're doing right now is setting up a new consortium. 543 01:02:43,400 --> 01:02:46,969 We got funded by the NIH, which was funded by The Age, 544 01:02:46,970 --> 01:02:55,040 to set up one of their avid drug discovery units, which is basically a virtual drug discovery unit. 545 01:02:55,280 --> 01:03:01,879 It's very well funded with $68 million across five years. 546 01:03:01,880 --> 01:03:04,200 And this is funding for pandemic preparedness. 547 01:03:04,670 --> 01:03:13,640 So basically generates so basically repeating what we've done for the moonshot, but all different kinds of viruses that may have pandemic potential. 548 01:03:13,910 --> 01:03:22,460 And we will reuse a lot of the kind of open access ideas and pro-competitive ideas for this new consortium. 549 01:03:22,820 --> 01:03:29,420 And that's that's a very interesting project for us because we can actually implement a lot 550 01:03:29,420 --> 01:03:34,050 of the learnings that we have from the moving into that similar drug discovery effort. 551 01:03:34,070 --> 01:03:37,070 I think there are lots of other diseases that. 552 01:03:38,630 --> 01:03:43,280 This could be relevant or parts of this could be relevant too. So I'm thinking. 553 01:03:43,310 --> 01:03:51,290 Amal So anti-microbial resistance would be one of the obvious fields where a lot of competitive science would be extremely 554 01:03:51,290 --> 01:03:59,180 worthwhile to the field and but also a lot of the kind of thoughts that we have on the kind of potential return on investment. 555 01:03:59,450 --> 01:04:09,290 Once you develop the drug, it is crucial to advance the field as well and potentially rare diseases. 556 01:04:09,290 --> 01:04:14,240 So, yes, there are things that we can learn. I don't think the exact setup could be repeated. 557 01:04:14,930 --> 01:04:22,940 That was probably something that was very new to the pandemic with the acceleration, even though we were much slower than, for example, 558 01:04:22,940 --> 01:04:27,310 Pfizer that was doing a very similar drug, although very different mechanism, 559 01:04:27,320 --> 01:04:32,660 like if they want to give it to that, we are having convey the point that they're having. 560 01:04:32,660 --> 01:04:38,210 So they looked at differences and. We couldn't have done it at that time scale. 561 01:04:39,230 --> 01:04:47,670 But and even our timescale is extremely accelerated in comparison to what might be possible in a non-pandemic setup. 562 01:04:48,680 --> 01:04:56,930 And also, I think the kind of demands and time contribution would be very difficult to generate in the future for other kinds of business. 563 01:04:56,930 --> 01:05:01,790 But I think a lot of the ideas should be taken forward in different avenues. 564 01:05:02,340 --> 01:05:05,660 Well, I mean. Oh, we've only got 3 minutes left. 565 01:05:10,820 --> 01:05:17,230 So which I have to pick a question to ask and. Well, last night, I guess the usual. 566 01:05:17,250 --> 01:05:19,350 Final question. So has the. 567 01:05:19,710 --> 01:05:27,990 Has your experience of working through this on this project through the pandemic changed your attitude or your approach to your work? 568 01:05:27,990 --> 01:05:30,570 And what would you like to change in the future? 569 01:05:34,300 --> 01:05:45,390 It has been absolutely crazy the last two years, both personally trying to kind of just get everything organised, 570 01:05:45,410 --> 01:05:50,560 like having the kids during the day, which you couldn't send back to school because we had COVID in the house. 571 01:05:50,800 --> 01:06:02,290 That would have been a key marker. And actually at some point Frank was admitted to hospital with COVID while the child was running. 572 01:06:03,820 --> 01:06:15,760 So I'm like both on it. I would like to to repeat the personal experience of the pure workload, but also the feel of urgency around work. 573 01:06:15,760 --> 01:06:23,860 So I'm actually looking forward to using all the bad eggs, the professional learnings that I had, 574 01:06:23,860 --> 01:06:31,240 and it has been an incredibly incredible learning experience and just thinking about forwards. 575 01:06:33,760 --> 01:06:44,980 Into into the next phase and using that to basically deliver compounds and and and knowledge for the common good. 576 01:06:45,700 --> 01:06:48,820 So that's that's that's something that I feel very passionate about. 577 01:06:48,840 --> 01:06:52,810 And I think I got even more passionate about it in the last two years. 578 01:06:53,230 --> 01:07:06,430 Seeing like how there is still such a divide between access to treatment and drugs here in the U.S. and kind of in the developed countries. 579 01:07:06,430 --> 01:07:15,399 And then compared to what what I can see and that starts with testing, actually, it's not actually just compounds with drugs. 580 01:07:15,400 --> 01:07:22,930 It's it's access to health care. So it's something that I feel even more passionate about than I felt beforehand. 581 01:07:22,990 --> 01:07:27,250 So that's something that is definitely more pronounced. 582 01:07:27,820 --> 01:07:37,090 What was absolutely unique and what I would like to carry forward as well is this feeling of We can do this together. 583 01:07:38,170 --> 01:07:44,450 And and that is something that I now also find very strongly in the pandemic preparedness. 584 01:07:44,450 --> 01:07:51,759 Feel that everything is working towards now is that we will need to keep this sense of togetherness. 585 01:07:51,760 --> 01:07:58,930 Some call to arms to be prepared for the next pandemic because the likelihood of that is unfortunately not that low. 586 01:08:00,490 --> 01:08:05,950 So that's that's something that I would like to maintain. 587 01:08:08,480 --> 01:08:12,800 Excellent. Thank you very much. Oh, I've lost my cursor again. 588 01:08:13,820 --> 01:08:16,220 Come on. Where is it? Thirties.