1
00:00:01,770 --> 00:00:06,090
Okay. Could you start by saying your name and your affiliation, sir?

2
00:00:06,210 --> 00:00:15,120
I'm Derek Cook and a consultant in infectious disease and clinical microbiology in the John Ratcliffe Hospital,

3
00:00:15,870 --> 00:00:20,250
and then also a professor of microbiology in the Buffalo Department of Medicine.

4
00:00:21,090 --> 00:00:27,720
Thanks very much. And just very briefly, can you give me some kind of account of how you got to be where you are now?

5
00:00:28,950 --> 00:00:35,520
But the first step in the journey perhaps was growing up in rural South Africa and

6
00:00:35,880 --> 00:00:42,870
very rural South Africa and then made my way to medical school in Johannesburg.

7
00:00:43,230 --> 00:00:49,680
And then from there went through the London School of Tropical Medicine and Hygiene in London and then

8
00:00:49,680 --> 00:01:00,000
did my specialist training in the United States and then came to Oxford for the city four years.

9
00:01:02,910 --> 00:01:06,220
And what made you pick? Microbiology, as I said.

10
00:01:07,450 --> 00:01:12,440
Well, it is more. Not microbiology so much as medicine.

11
00:01:13,450 --> 00:01:16,690
I trained as a physician transmits infectious disease.

12
00:01:18,260 --> 00:01:25,999
Dr. And I've done a lot of general medicine and infectious disease work, but also trained in pathology,

13
00:01:26,000 --> 00:01:33,320
sort of became qualified as what one would call a jobbing microbe ologist.

14
00:01:33,980 --> 00:01:45,380
But preceding that, I've done a lot of research in the United States on microbial behaviour and that was.

15
00:01:46,580 --> 00:01:53,879
Sort of. A natural a starting point to do what we refer to as clinical microbiology.

16
00:01:53,880 --> 00:01:58,500
And the differentiation between them at many levels is not great to enough.

17
00:01:58,500 --> 00:02:05,760
Two other differences quite substantial living across the practise of medicine but infectious disease,

18
00:02:05,760 --> 00:02:10,020
but also having an understanding of the scientific basis of the microbes that cause disease.

19
00:02:11,460 --> 00:02:16,560
So summarising it, I mean, the question I've written down is I'm asking everybody this.

20
00:02:16,740 --> 00:02:22,080
If you had to define a single big question that gets you out of bed in the morning, what would it be?

21
00:02:22,590 --> 00:02:26,670
But I could rephrase that. I'm not sure if it's the same question or not as what appeals to you about bugs.

22
00:02:33,380 --> 00:02:40,550
That's sort of an interesting question because it's it's really solving problems that gets me out of bed in

23
00:02:40,550 --> 00:02:47,440
the morning rather than and tax bucks represents one end of a spectrum of troublesome problems that you can,

24
00:02:47,820 --> 00:02:58,830
can, can. Entertain yourself with a little and excite yourself with other levels and the diseases that cause a

25
00:02:58,830 --> 00:03:04,890
fascinating in the way that they interact with their hosts and affect hosts at an individual level,

26
00:03:04,900 --> 00:03:06,270
at a population level.

27
00:03:07,020 --> 00:03:17,940
So it's really probably the interface between medicine and biology and all of that through fields that's appealing and interesting.

28
00:03:19,170 --> 00:03:22,229
I like caring for patients. I like solving problems in the lab.

29
00:03:22,230 --> 00:03:27,900
I like writing papers. I'm lucky to do a lot of those.

30
00:03:30,180 --> 00:03:39,060
So let's move on to the pandemic. Can you remember how you first heard that there was a pandemic in the offing?

31
00:03:41,540 --> 00:03:46,680
Well, I mean, it was just in the media and it was a. A.

32
00:03:49,350 --> 00:03:56,760
A reference to it in the media, which was enormously similar to the first source outbreak.

33
00:03:56,910 --> 00:04:03,500
It is completely reminiscent of that giant January or was out before it was December.

34
00:04:03,510 --> 00:04:13,820
It was December 2019 and 19 December who had and I don't know whether 12 or 14 years before that.

35
00:04:15,310 --> 00:04:29,860
The settlers outbreak that started in the southern provinces of China and then became prevalent in Hong Kong and didn't become a pandemic.

36
00:04:31,380 --> 00:04:34,810
Well, this one did. But they're very similar at the beginning.

37
00:04:35,080 --> 00:04:45,180
Yeah. And did you how did you sort of slow realisation that this was something that was likely to affect us?

38
00:04:45,180 --> 00:04:51,960
Or did you initially think, Oh, that's interesting, there's another cell is happening in the forest and not expect it to?

39
00:04:53,280 --> 00:04:56,300
I didn't have very strong expectations one way or another.

40
00:04:56,310 --> 00:05:02,070
I mean, so as the first epidemic petered out, the Zika.

41
00:05:03,420 --> 00:05:07,479
If it was huge in the media.

42
00:05:07,480 --> 00:05:14,610
But for editing that was in Brazil, it's well, it was one of the particular first.

43
00:05:14,610 --> 00:05:18,760
So I was in Brazil for it preceded it, I think it was evident before.

44
00:05:18,910 --> 00:05:24,390
So I grew up with Zika in Africa. So big deal and.

45
00:05:25,900 --> 00:05:41,350
And then there were similar limited outbreaks of of of other viral illnesses, such as sort of curvy in in the Middle East.

46
00:05:41,410 --> 00:05:44,890
And there's obviously a particular outbreak in South Korea.

47
00:05:45,400 --> 00:05:49,900
So you wouldn't was aware of this. It was only really.

48
00:05:53,190 --> 00:06:01,620
January, February. Did you realise that there were going to be problems in an office of the W.H.O. in.

49
00:06:03,810 --> 00:06:11,280
Like January. And it was just becoming apparent that Italy, the north of Italy, was having a particular problem.

50
00:06:12,270 --> 00:06:19,140
And then it pretty rapidly became clear that this was going to spread around the world.

51
00:06:22,470 --> 00:06:32,230
I was of first. I feel particularly alarmed about it because it was going to do from pretty even temper tantrums.

52
00:06:33,760 --> 00:06:39,010
I looked at it with equanimity rather than a huge mess.

53
00:06:39,470 --> 00:06:45,430
Mm hmm. But did you think this is going to change the way I work or what I work on?

54
00:06:46,600 --> 00:06:53,020
No, it isn't going to change the way that that plan would work at that stage.

55
00:06:53,920 --> 00:07:03,850
But it soon became clear that it was going to be a need for a huge effort and a massive response.

56
00:07:06,630 --> 00:07:11,820
That's what we've vaguely anticipated, what would be involved in that?

57
00:07:12,060 --> 00:07:22,410
But in other parts of what used to be called Zulu land in South Africa, what about other cholera outbreak outbreaks?

58
00:07:23,340 --> 00:07:29,010
I've directed a major part of the Public Health England where we were thinking about these sorts of things.

59
00:07:32,060 --> 00:07:39,200
I'm not taken with a long speech. I'm not trying to alarm the children.

60
00:07:39,300 --> 00:07:49,760
I mean. But. Yes, yes, yes. I was matter of fact about it. If if it grew into something that was huge, one would deal with it and become involved.

61
00:07:50,750 --> 00:07:55,380
So just this is just sort of by the way, that's always been the case. It wasn't just because this was about you.

62
00:07:55,450 --> 00:08:00,440
Yeah. Yeah. So just to my background, to what we can talk about later.

63
00:08:00,830 --> 00:08:12,320
Can you just outline briefly the natural history of the the epidemic as it's evolved from from December 2019 till till now,

64
00:08:12,650 --> 00:08:17,960
in terms of where different variants appeared around the world and how they've affected case numbers.

65
00:08:18,110 --> 00:08:34,850
So yeah, there's some aspects of it that were pretty predictable in the sense that if you have a pandemic, it's just around the world.

66
00:08:35,480 --> 00:08:41,680
You knew that it would go away. So that was very clear.

67
00:08:41,680 --> 00:08:49,670
It occurred ways that because of the immunity arising and then very people don't really know why waves happen.

68
00:08:50,570 --> 00:08:58,170
Flu does. This is the biology behind this sort of not that clear.

69
00:08:59,100 --> 00:09:03,710
I mean, the 1918 flu epidemic were.

70
00:09:05,000 --> 00:09:11,780
Please. It's interesting. Must be a bit of you being a part of the way people behave.

71
00:09:12,800 --> 00:09:17,000
But this season. A lot of variables.

72
00:09:17,690 --> 00:09:24,410
So that was all going to be expected in my mind.

73
00:09:25,250 --> 00:09:32,660
What wasn't really prepared for is how the properties of the virus would change.

74
00:09:33,900 --> 00:09:44,580
And they have changed to list some characteristics. It is, in retrospect, easy to understand so that it would be completely intuitive.

75
00:09:45,600 --> 00:09:49,400
But you didn't expect it was going to happen.

76
00:09:49,410 --> 00:10:00,270
And those that those changes had to do with how to adapt it to the human which wasn't expected or other animals, as it did to me.

77
00:10:02,340 --> 00:10:05,910
But it would become hugely more transmissible.

78
00:10:07,110 --> 00:10:17,249
And then a lot of the hysteria increase of pathogens of the species is that it could be more

79
00:10:17,250 --> 00:10:25,650
variant at the beginning and over time adapt to the host and become a lot less virulent.

80
00:10:26,430 --> 00:10:34,290
Not clear that's happened yet, but one of the reasons for that immune escape is very well recognised.

81
00:10:35,620 --> 00:10:43,350
So that wasn't really a surprise. The stick changes and transmission was a bit of a surprise.

82
00:10:43,590 --> 00:10:48,210
This is with the emergence of new variants. It's only son.

83
00:10:49,470 --> 00:10:53,070
So you mean the same variant can increase its transmissibility?

84
00:10:55,510 --> 00:11:01,360
Well, what you're hitting on is something that got to do is lazy terminology.

85
00:11:03,520 --> 00:11:10,060
Variances in some settings are very, very specific terms.

86
00:11:10,120 --> 00:11:15,040
You get a single alteration in the genetic makeup.

87
00:11:15,520 --> 00:11:22,600
But in terms of the way one has used it for the pandemic is.

88
00:11:24,180 --> 00:11:30,780
Actually a specific word for it would be a lineage.

89
00:11:31,840 --> 00:11:40,730
It's quite often has a number of mutations in it. And then in respect of transmissibility, it's.

90
00:11:41,900 --> 00:11:48,170
It's a causative mutation that occurs, occurs and that can be on lots of different individuals.

91
00:11:49,430 --> 00:11:54,590
And therefore variant is a funny way of describing it.

92
00:11:54,800 --> 00:12:00,620
You might be aware of that, but it's it's confusing to to a lot of people.

93
00:12:00,620 --> 00:12:10,370
And the more that the virus combines with itself or with it, you get kids get convergent evolution.

94
00:12:10,370 --> 00:12:16,110
That's where the same mutation occurs and very different. Settings are linked.

95
00:12:17,140 --> 00:12:33,370
Settings of the virus because of selection, you rapidly realised that the business was down to a few specific mutations and that's been surprising.

96
00:12:33,520 --> 00:12:42,430
The extent to which the transmissibility of the virus has been changed by some specific mutations or might be a combination of mutations, may vary.

97
00:12:42,940 --> 00:12:45,379
Researchers, not myself,

98
00:12:45,380 --> 00:12:52,840
but researchers are very interested in how that might be additive with synergistic effects between combinations of annotations,

99
00:12:52,840 --> 00:13:01,930
but it's certainly mostly down to single mutations, resulting in these quite profound changes in the behaviour of virus.

100
00:13:03,470 --> 00:13:17,480
And some of them are thought to go hand in hand with altering it to the ability of the virus to escape immune protection of the host.

101
00:13:19,580 --> 00:13:22,640
Which was expected. I mean, that is quite common.

102
00:13:24,140 --> 00:13:27,410
So one thing I should have asked you earlier is I've I've got a question here.

103
00:13:27,440 --> 00:13:30,200
What methods do you use to explore the question?

104
00:13:30,220 --> 00:13:37,940
And I know you've been involved in a project, you know, if it's a project, an umbrella thing called modernising medical microbiology.

105
00:13:38,760 --> 00:13:43,700
So can you just explain for the uninitiated what that involves?

106
00:13:46,720 --> 00:13:56,030
So that that collection of people that have come together in this modernising medical microbiology arose from a large quantity.

107
00:13:56,030 --> 00:14:01,690
We succeeded in getting. And it's it's it's grown organically.

108
00:14:02,140 --> 00:14:05,320
Independent research is working very closely together.

109
00:14:06,070 --> 00:14:09,160
And that's something a number of years ago.

110
00:14:09,430 --> 00:14:12,459
In 2008. Yes. Yes.

111
00:14:12,460 --> 00:14:21,400
That's about the. And. The ambition there was to.

112
00:14:23,060 --> 00:14:28,370
Make use of new technology. So we're going to bear on.

113
00:14:30,570 --> 00:14:38,430
Science in general, but in particular on microbial science and human infectious disease science.

114
00:14:38,730 --> 00:14:44,640
And you could look at them holistically rather than as discrete parts of it.

115
00:14:45,330 --> 00:14:50,040
And the technologies that are driving that.

116
00:14:51,720 --> 00:15:01,040
Digital computational. Technology is quite obviously and very closely linked to that to the.

117
00:15:02,240 --> 00:15:09,800
Tools enable you to recover the genetic code and in its entirety.

118
00:15:11,060 --> 00:15:16,730
And the synthesis of those two made you able to assemble.

119
00:15:18,420 --> 00:15:34,860
The code that's generated in little bits to form the whole chromosome with the genetic structure and then to mine that reflation again.

120
00:15:36,400 --> 00:15:44,330
This. It's highly dependent on really advanced computational processes.

121
00:15:44,330 --> 00:15:49,310
It's really fancy software, very high performance computers.

122
00:15:50,480 --> 00:15:56,140
And then if you have that layer of information, how can you link it?

123
00:15:56,720 --> 00:16:04,070
You know, if you're doing medical science to people and to do things like tracking infection in hospitals.

124
00:16:04,680 --> 00:16:14,500
Yes. You know, understanding all manner of things, what the risk factors that account for people developing disease.

125
00:16:14,580 --> 00:16:21,260
You could do all of the same things with human genetic code and it was more microbial in.

126
00:16:21,260 --> 00:16:25,819
But there's quite a strong piece that's got to do with human genomics,

127
00:16:25,820 --> 00:16:32,690
but a huge part that's got to do with large datasets about humans and health care systems.

128
00:16:34,040 --> 00:16:44,839
What do people with a certain infection get? Severe illness and some very deadly illness, and how does that distribute across populations?

129
00:16:44,840 --> 00:16:47,990
And to do those very large.

130
00:16:49,790 --> 00:16:54,650
Analysis across large numbers of people and the microbes that they've got.

131
00:16:55,490 --> 00:16:58,580
You could argue address by having confidence computing again.

132
00:16:58,670 --> 00:17:02,760
I mean, it's very robust information. You're going to speak to Sarah Walker.

133
00:17:02,780 --> 00:17:08,060
She's become absolutely key in that sort of work.

134
00:17:09,320 --> 00:17:17,209
And and that was really the background thinking about modernising medical microbiology schemas.

135
00:17:17,210 --> 00:17:24,140
How can you take these new technologies and as a result of studies, generate new information,

136
00:17:24,560 --> 00:17:29,960
identify better ways of doing things that include better diagnostics?

137
00:17:30,710 --> 00:17:37,130
And, you know, how can you identify ways of of of.

138
00:17:40,070 --> 00:17:48,080
Recognising the best ways of preventing the spread of infection or even recognising when you have spread of infection, you can have interventions.

139
00:17:49,250 --> 00:17:58,280
I think we have had a marvellous time, you know, again, holistically looking at the subject.

140
00:17:59,240 --> 00:18:03,830
Can you give me some examples of some of the kinds of infections that you were interested in before COVID came along?

141
00:18:05,120 --> 00:18:21,260
Well, a couple that have been highly illuminating if you take a disease caused by Clostridium difficile.

142
00:18:22,980 --> 00:18:26,389
We had a. Amongst us.

143
00:18:26,390 --> 00:18:28,390
We got a sense of humour, decency.

144
00:18:30,140 --> 00:18:53,850
There was a great political imperative, almost interest, in ascribing the presence of these jobs as the disease to dirty hospitals.

145
00:18:53,870 --> 00:18:57,140
I don't know if you can remember it in the press a lot.

146
00:18:57,740 --> 00:19:11,819
We were sort of. Kind of unimpressed with reset and set about very deliberately to investigate the organism with that

147
00:19:11,820 --> 00:19:20,670
particular question in mind and did a huge amount of work very carefully and in a systematic way,

148
00:19:20,670 --> 00:19:27,120
collecting cases with the germ and with information about them and the germs themselves,

149
00:19:27,120 --> 00:19:39,269
and were able to recover the genetic code by sequencing the microorganisms and then did combined epidemiological and microbial investigations.

150
00:19:39,270 --> 00:19:50,810
The. The outcome of which was that the evidence for the hospital being dirty and the source of the infection was by mute.

151
00:19:52,250 --> 00:20:01,490
And it was much more of a matter of these organisms increasing into the hospital environment, probably through other people.

152
00:20:02,930 --> 00:20:06,260
And so people could carry them around and be perfectly well.

153
00:20:06,750 --> 00:20:07,980
I mean definitely, yeah.

154
00:20:08,000 --> 00:20:18,740
It's like 98, 99% of the germs life is in people who don't even know they've got it, never had a problem with it, but maybe even 99.5%.

155
00:20:21,710 --> 00:20:29,330
Particularly children. Write another story. And that was sort of entertaining not to be opposed.

156
00:20:29,330 --> 00:20:37,350
The idea of keeping the hospitals, but the idea that the germ represented a market for a dirty hospital per se.

157
00:20:37,840 --> 00:20:41,080
It was a marker for a lot more than that huge amount.

158
00:20:41,120 --> 00:20:52,550
Not all of it. Entirely teasing, and then we were able to demonstrate pretty convincingly.

159
00:20:55,290 --> 00:21:01,799
Better than pretty convincingly that the whole so-called epidemic had to do with the way

160
00:21:01,800 --> 00:21:08,190
that germ had become resistant to an antibiotic called ciprofloxacin quinolone antibiotics.

161
00:21:08,760 --> 00:21:21,959
And what you had is the linkage of Clostridium difficile that acquired this causative gene that made it resistant to quinolones.

162
00:21:21,960 --> 00:21:27,270
And then it's very rapidly. True selection.

163
00:21:28,180 --> 00:21:33,460
But black sites could be two reasons why.

164
00:21:34,570 --> 00:21:49,060
This is the selection going on in the background. And once Quinolone usage went down some 50% in the community amongst general practitioners.

165
00:21:50,150 --> 00:21:58,900
These lineages reduced in number and disappeared and the epidemic disappeared with it.

166
00:21:59,260 --> 00:22:06,430
So it had nothing to do with the hospitals and had everything to do with the use of these quinolone antibiotics.

167
00:22:07,240 --> 00:22:15,400
So that was sort of entertaining and a counter to various political slogans.

168
00:22:15,940 --> 00:22:28,900
We enjoyed that some loved and that you could never have discovered that without extremely sophisticated computing and DNA sequencing and systematic,

169
00:22:29,140 --> 00:22:34,120
careful accumulation of information about the patients,

170
00:22:34,120 --> 00:22:38,270
which we got from linking clinical records systems in the hospital and getting all

171
00:22:38,290 --> 00:22:44,010
the the necessary information governance arrangements organised by the hospitals,

172
00:22:44,020 --> 00:22:54,670
a huge effort. All those permissions to do it in a very diligent and safe way in terms of I think we've done some of the work with TB.

173
00:23:07,980 --> 00:23:12,780
With TB. It's got lots of dimensions to it,

174
00:23:12,780 --> 00:23:27,930
but one of the most pleasing is that we can take the genetic code and mine the genetic code and find signatures of resistance in the genetic code.

175
00:23:29,430 --> 00:23:44,610
Get closer to mutations. There's a theme here and they they can be one amongst 4 million, 400,000 other nucleotides.

176
00:23:44,910 --> 00:23:50,670
And that one change turns the organism from susceptible to resistant.

177
00:23:51,210 --> 00:24:03,750
If you can very reliably identify those changes and determine purely genomically whether the organism is going to be resistant to.

178
00:24:05,840 --> 00:24:17,530
Activity, licence, drug or not, and make decisions about whether to use it without doing the tedious job and the big faff of growing

179
00:24:17,530 --> 00:24:23,530
its growing in the presence of the drug and figuring out if it grows from what started when we just

180
00:24:23,620 --> 00:24:29,979
completed a huge global project across 27 countries around the world doing this in scale and then just

181
00:24:29,980 --> 00:24:40,420
publishing a large number of patent papers that enable one to make those deterministic predictions from.

182
00:24:42,440 --> 00:24:48,950
Computationally analysing the virtual codes and singling out which drugs that particular

183
00:24:48,950 --> 00:24:54,020
strain would be resistant to and whether they are susceptible to those drugs.

184
00:24:54,060 --> 00:24:57,590
And you can make that differentiation with a very high degree of accuracy.

185
00:24:58,840 --> 00:25:02,830
And it's transmogrified how you practise.

186
00:25:04,250 --> 00:25:09,840
You know. Tbe Manager management.

187
00:25:10,610 --> 00:25:14,300
And it was very heavily used.

188
00:25:15,540 --> 00:25:21,450
In assisting the W.H.O. created a catalogue, as they call it,

189
00:25:21,690 --> 00:25:27,120
which is just a repository of the information about each one of those changes in the

190
00:25:27,120 --> 00:25:33,840
genetic code so that people could develop tests for this various configurations and.

191
00:25:35,150 --> 00:25:43,190
We really did most of the work towards creating this W.H.O. dust catalogue.

192
00:25:43,610 --> 00:25:51,620
You think it could be a lot better when you're publishing all the papers too, to understand and target you would do that.

193
00:25:51,910 --> 00:25:55,280
So again, that is the computation.

194
00:25:55,730 --> 00:26:01,610
Huge, systematically collected samples around the world, brought together and analysed.

195
00:26:01,620 --> 00:26:12,670
And it's not just medicine scientists of all different sorts as people from all around the world that want to effectively work together.

196
00:26:12,680 --> 00:26:22,200
We had fantastic success working with China, India, the United States, and then many countries in Africa, Asia.

197
00:26:22,820 --> 00:26:35,750
South America. So just to drag you back to Kobe, you had this tremendous set of tools for interrogating pathogens, infectious disease pathogens.

198
00:26:36,440 --> 00:26:49,880
What point did you decide that you were going to turn some of that power on this new, new infection and help us do it with different drugs?

199
00:26:50,270 --> 00:27:00,190
Well, how did it just what how did it happen? Well, as Sara and I said, you know, this is going to be a ridiculous feeding frenzy.

200
00:27:00,260 --> 00:27:04,880
But to do research, we'll just work to support the general effort.

201
00:27:05,060 --> 00:27:11,010
Right. And then about. February.

202
00:27:11,020 --> 00:27:16,590
March. John Bell contacted me.

203
00:27:16,620 --> 00:27:22,490
He said, you know, in government we've got a huge problem because we can't find people to.

204
00:27:23,830 --> 00:27:29,530
Do what would be considered banal, menial, almost menial jobs.

205
00:27:30,400 --> 00:27:33,490
Trying to stand up huge scale testing in the country.

206
00:27:34,600 --> 00:27:41,410
And we don't even know whether any of the techniques involved in testing just how you swap or troops you carry stuff in,

207
00:27:41,890 --> 00:27:51,850
how you process through through PCR machines, whether any of them meet the standards means we can roll this out in the country.

208
00:27:53,680 --> 00:28:07,569
And then they were starting to market a whole lot of kids to see whether you picking your finger had antibodies to to the virus indicating

209
00:28:07,570 --> 00:28:19,360
that you you had some element of immunity and there were all manner of people trying to sell government's complete rubbish and said,

210
00:28:19,540 --> 00:28:22,750
we know this. Can you do stuff to help us set it all up?

211
00:28:23,100 --> 00:28:29,319
We will we will assist in the effort.

212
00:28:29,320 --> 00:28:36,410
So we set up a programme of work between here and at Porton Down and our new people from Porton Down.

213
00:28:36,410 --> 00:28:43,180
So to Public Health, England, that started screen swaps,

214
00:28:43,180 --> 00:28:50,750
just standard swaps to make sure that they were safe to use and it would not affect the testing values.

215
00:28:50,830 --> 00:28:57,670
And we did talk about 500 swaps, but just the actual swap, just to stick with the bits of cotton wool to make sure it works.

216
00:28:57,670 --> 00:29:03,040
Right. My goodness, we didn't get hundreds because we think they're going to be manufactured.

217
00:29:03,040 --> 00:29:09,810
That would be flown into the country that were all the supply problems and you didn't know if you were buying rubbish or things that were.

218
00:29:10,300 --> 00:29:17,260
So we set this up on a massive scale so they could buy with the understanding or knowledge that

219
00:29:17,440 --> 00:29:24,990
these standards that you would say of safe to use and would not alter the performance of tests.

220
00:29:25,150 --> 00:29:28,090
That's what it was, but no mindless sort of work.

221
00:29:28,160 --> 00:29:36,370
We set up systems to do it on a huge scale and then we did the same to Immunity and I managed to call up friends around the

222
00:29:36,370 --> 00:29:45,609
country because people had been infected with and then we had these banks of Sierra that we could then test and rollout tests.

223
00:29:45,610 --> 00:29:49,600
We developed a huge allies of test here, just expanding the likes of test.

224
00:29:50,320 --> 00:29:53,500
So it's a particular way that you measure antibodies.

225
00:29:54,520 --> 00:29:58,180
So this was invented many years ago.

226
00:29:59,110 --> 00:30:02,230
So antibody target, something which you can say is an antigen.

227
00:30:02,530 --> 00:30:16,030
If you mobilise antigen somewhere, you can bring the antibody solution close to the localised antigen, targeted binds to the target.

228
00:30:16,540 --> 00:30:23,529
And then you have ways of detecting the presence of the antibody on the antigen target because it binds very tightly.

229
00:30:23,530 --> 00:30:33,759
If you wash stuff away, you're left with the antibodies and you can get signals for its presence and you can put this into a whole range of settings.

230
00:30:33,760 --> 00:30:41,440
And we did this through Clayton and I think you place it through a pretty to trace you've seen them but how does that

231
00:30:41,440 --> 00:30:50,590
differ from the commercial antibody tests of which I've done one because of the UK Biobank them to finger prick ones.

232
00:30:51,160 --> 00:30:57,910
It's a different way of presenting the antigen and operating the test.

233
00:30:58,090 --> 00:31:03,790
They're all the same biochemical reaction behind it, but different representations of it.

234
00:31:04,210 --> 00:31:20,970
So we've got the national testing for the COVID 19 epidemiological survey is run by the Office of National Statistics.

235
00:31:20,980 --> 00:31:27,550
I can take you ensure that we run through 5000 to 6000 samples today just around the corner here

236
00:31:28,150 --> 00:31:36,700
and test them on machines for the presence of antibodies to different parts of the virus spike

237
00:31:36,700 --> 00:31:46,179
apart from the protein part and the beginning of the journey to doing that started with being

238
00:31:46,180 --> 00:31:52,180
able to do the testing to demonstrate these various kinds of people have made we're ready,

239
00:31:52,180 --> 00:32:03,690
almost uniformly hopeless. And on the side we started to make one that would perform to hugely improve support you with performance drugs.

240
00:32:04,570 --> 00:32:18,330
And now we've got that now fully established as a test with all the regulatory approvals that run 150,000 tests a month, and it's absolutely not that.

241
00:32:18,380 --> 00:32:21,460
So we're telling you because that's had to be used in a lab setting.

242
00:32:21,940 --> 00:32:26,900
Can that. Yes. Yes, yes, yes, yes.

243
00:32:27,050 --> 00:32:36,120
If you get 6000 samples coming to the hospital for this is more than the entire pathology service in this hospital runs a day.

244
00:32:36,450 --> 00:32:40,830
Right. So you've got this. I'll show it to you. It's stuck.

245
00:32:41,010 --> 00:32:42,270
My gosh. How did that happen?

246
00:32:43,950 --> 00:32:55,589
And so and that work was going on very closely to Jordan, working with the government, trying to stamp things up and end it.

247
00:32:55,590 --> 00:32:58,649
But at some point in this process, when it was quite early on, wasn't it?

248
00:32:58,650 --> 00:33:03,240
It was early March when the government suddenly said, we're going to have to stop testing.

249
00:33:04,290 --> 00:33:09,490
So we were doing the PCR testing. That was accuracy.

250
00:33:10,710 --> 00:33:22,950
It's actually that's quite interesting. That was the huge. Sort of realisation that Public Health England wasn't able to deliver the services that

251
00:33:22,950 --> 00:33:31,560
would enable testing and many of the the the components of the national response.

252
00:33:32,740 --> 00:33:37,000
In fact, I think the organisation went through a minor collapse.

253
00:33:37,150 --> 00:33:46,270
That's how we got involved. They couldn't process these samples within their management structure because we put them down,

254
00:33:46,330 --> 00:33:52,360
because we use the people that we set up a system for doing it with very able people.

255
00:33:52,360 --> 00:33:58,090
But the formal structures that they had, the response malfunctioned early on.

256
00:33:58,660 --> 00:34:02,440
And one of the aspects of that was working with the NHS.

257
00:34:03,040 --> 00:34:13,720
They weren't able to run our tests and that really was a huge interest in setting up these lighthouse labs and mega labs.

258
00:34:13,840 --> 00:34:15,700
The decision was highly controversial.

259
00:34:15,710 --> 00:34:23,680
It wasn't a very straightforward thing and there was a lot of rancour around it, like a local hospital was very upset about it, hugely upset about it.

260
00:34:24,670 --> 00:34:34,570
But to do it on scale, you had to deliver the reagents and the tools to do it, swabs and all the rest.

261
00:34:35,140 --> 00:34:41,020
And you had to stand up to the labs themselves, which we didn't really get directly involved in,

262
00:34:41,020 --> 00:34:46,389
although we put a huge effort into making sure that the workflows functioned very well,

263
00:34:46,390 --> 00:34:50,170
the tabulate tested those and put them on to the right footing.

264
00:34:51,040 --> 00:34:55,150
So and I got to know some of those people very well.

265
00:34:57,120 --> 00:35:02,640
As well as setting up things in the hospital because we have a responsibility and a job to do that locally.

266
00:35:02,910 --> 00:35:07,830
So this wasn't really research. No, but it was a COVID response.

267
00:35:07,830 --> 00:35:16,020
So just so it counts case. And so all those wonderful things we were doing and then we have lots of people around.

268
00:35:16,020 --> 00:35:27,240
They were enthusiastic. We came to work every day. We worked, you know, a minimum of ten, 12, 14 hours every day, non-stop, all three weekends.

269
00:35:27,540 --> 00:35:33,779
And you say it wasn't research, but there were papers published. I mean, I've seen the paper in which you published this.

270
00:35:33,780 --> 00:35:37,590
How, how rubbish the commercial antibody test.

271
00:35:38,090 --> 00:35:48,030
We published more stuff this year. So we, we drew together all the young researchers we have all the old ones as well,

272
00:35:48,120 --> 00:35:54,780
everybody and we both teams, we've worked unbelievably and unrelentingly through this whole process.

273
00:35:55,650 --> 00:36:04,280
And then gradually as we and then Sara got recruited to go and run or in this study as the chief investigator.

274
00:36:05,130 --> 00:36:08,860
In fact, she sits next to her. So she she lives and breathes.

275
00:36:11,220 --> 00:36:16,590
And then we started to build very substantial research,

276
00:36:16,590 --> 00:36:22,560
focussed around along the lines of we spoke about getting back to the genomic kind of research.

277
00:36:23,460 --> 00:36:35,460
We are working on that very intensively as a consequence of a charitable philanthropic donation from Article ten years worth of their cloud resources.

278
00:36:38,410 --> 00:36:54,320
Unlimited to support. Genome sequence data assembly variant coding mutations emit another analysis storage around the world.

279
00:36:54,500 --> 00:37:01,880
So this is this builds on your scalable pathogen pipeline platform.

280
00:37:02,960 --> 00:37:09,500
And this is not all that's about that. Well, that was a very lovely little innovation that was made.

281
00:37:11,790 --> 00:37:13,470
Out of tbe work we did.

282
00:37:13,800 --> 00:37:33,180
We took the first very poorly designed, very poorly coded software for taking the sequences of TB, assembling them and identifying other.

283
00:37:34,600 --> 00:37:39,610
Samples of TB the group very closely related and we could.

284
00:37:41,350 --> 00:37:50,710
Quantify that relationship so that we know that that been recently acquired from the person.

285
00:37:53,200 --> 00:38:01,420
For one of two people who had it, if you reduced it down to two, was a cluster that would require that somehow from each other.

286
00:38:02,170 --> 00:38:10,720
And that became this unit one known for identifying outbreaks of TB and is known nationally and how you trace those people.

287
00:38:12,610 --> 00:38:21,040
Test and trace and then how you could predict resistance and all of that work.

288
00:38:21,040 --> 00:38:30,990
We were able to say just from the gene. Code analysis that you've got to suspect that one can use any one of the drugs safe.

289
00:38:31,410 --> 00:38:35,440
You know, you've got a resistant one. Think about that a bit more, which we can do.

290
00:38:35,440 --> 00:38:41,710
And that's a national service. We developed that it became very clear to us that we had to.

291
00:38:44,170 --> 00:38:50,499
Change the paradigm and that actually we wouldn't be able to establish with

292
00:38:50,500 --> 00:38:55,810
computational resources locally and many data centres or computer clusters,

293
00:38:56,170 --> 00:39:07,159
we'd have to put the software to the cloud because it became very clear that we couldn't design and create software that reached industry standards.

294
00:39:07,160 --> 00:39:13,899
So we started working very hard on trying to understand what you could do better and hoped that we

295
00:39:13,900 --> 00:39:24,070
would eventually have a partner with some tech organisations that would aid and assist in doing this,

296
00:39:24,070 --> 00:39:29,920
as you would say properly. And that has happened as a result of the pandemic.

297
00:39:30,670 --> 00:39:36,010
And it was recognised that the code that we had written that would provide a

298
00:39:36,010 --> 00:39:41,800
platform for running the tools could be run in any one of the big global clouds.

299
00:39:43,920 --> 00:39:52,850
And that was called this piece. I don't know what it is, but it's a scalable pathogen pipeline platform.

300
00:39:55,170 --> 00:40:00,450
And that was it. That's what the West is. You know, it's in the name but yet did.

301
00:40:01,680 --> 00:40:09,780
And that was recognised as a as a. That's quite an innovative development.

302
00:40:10,200 --> 00:40:14,460
The sad thing is that as innovative as it was,

303
00:40:14,550 --> 00:40:23,400
it's not scalable because there's so many other attributes that go into designing something like that and coding it that make it scalable,

304
00:40:23,400 --> 00:40:34,020
but which you mean you can put it on these cloud technologies and you can run in parallel, not just ten,

305
00:40:34,020 --> 00:40:41,520
not just 100, but thousands and tens of thousands of these jobs simultaneously and bring them together.

306
00:40:42,030 --> 00:40:49,320
So that requires coding and platform, that enables that.

307
00:40:49,830 --> 00:40:53,580
And although this intended to do it, it didn't do it very efficiently.

308
00:40:54,270 --> 00:40:58,110
It's interesting to see that of the experienced Oracle.

309
00:40:58,110 --> 00:41:01,799
Now that we are learning that you're going to have to redesign a lot of that.

310
00:41:01,800 --> 00:41:07,390
You have to refactor it. And you have to. Restructure a great deal.

311
00:41:07,390 --> 00:41:10,660
And so we had the insights before the pandemic.

312
00:41:10,660 --> 00:41:19,900
The pandemic came along. We can now do the thing that we think to do much better, massively, massively that we would have otherwise done.

313
00:41:20,230 --> 00:41:30,880
So what? What are the questions that you can answer using that that particular technology in relation to the pandemic or in relation to this virus?

314
00:41:33,420 --> 00:41:40,260
I don't think it's this is incremental. This is a massive step change in terms of what you can answer.

315
00:41:40,800 --> 00:41:44,400
What it is a step change in is accessibility to it.

316
00:41:44,630 --> 00:41:51,930
All right. Induction of cost. And it turns it into an everyday commodity for that and anywhere in the world.

317
00:41:52,160 --> 00:41:57,420
Yes, it becomes agnostic. You don't have to have expert people to take.

318
00:41:58,080 --> 00:42:09,350
It's taking. You know, it's taking A to Z out of London and put it into Jeeps.

319
00:42:10,310 --> 00:42:15,770
That's what this is doing. So it's taking what a lot of people loved.

320
00:42:15,770 --> 00:42:21,940
It's having something you can have a hold and feel and look at and understand the concern.

321
00:42:22,810 --> 00:42:25,940
I could be rude about it and turn it into something.

322
00:42:25,940 --> 00:42:32,050
But this is digital, this intuitive. Ooh. That's touching point.

323
00:42:32,290 --> 00:42:42,100
So you've got people in Africa, Philippines, everywhere collecting samples and being able to extract it, to extract the DNA, presumably.

324
00:42:42,220 --> 00:42:44,410
Yeah. There's a lot of yeah, a lot of hard stuff to do.

325
00:42:44,410 --> 00:42:52,810
But once they've got a genome sequenced, the challenge was to take the genome sequence, as it implied, already assemble it.

326
00:42:54,130 --> 00:43:02,360
And then once you've assembled mining for the relevant pieces, after you've got the complete assembly in place.

327
00:43:02,830 --> 00:43:04,060
And then having done that,

328
00:43:04,210 --> 00:43:12,070
that you can analyse it and then tested to demonstrate that it's performing in the way that you want to perform against some standard.

329
00:43:12,880 --> 00:43:19,720
And we can do that now so that somebody's got this type of secrets or that type of

330
00:43:19,720 --> 00:43:25,990
sequence or this kind of method of preparing it and upload the information to the cloud.

331
00:43:25,990 --> 00:43:35,200
And it comes back with the answers. And that hasn't been possible without having an expert locally to do all of that stuff,

332
00:43:35,940 --> 00:43:41,980
work on the command line computer and even write a bit of code to make it work.

333
00:43:42,580 --> 00:43:47,870
This is. At Europe's. You want to go somewhere?

334
00:43:48,650 --> 00:43:52,129
That's how you go. That I've got this DNA sequence.

335
00:43:52,130 --> 00:44:00,770
What is it? Search thoughts. So the Oracle people are doing that work with with you, with group that we're providing.

336
00:44:01,790 --> 00:44:08,630
It's a partnership, actually. There's a whole lot of things that they do unbelievably better than we can never do software as well as they can.

337
00:44:09,290 --> 00:44:14,000
Although we are establishing a vehicle to address that.

338
00:44:14,750 --> 00:44:22,879
They donate T to Oxford to provide the tools that enables you to do those same

339
00:44:22,880 --> 00:44:30,170
decent analysis that they will provide the infrastructure on there in the cloud to

340
00:44:30,170 --> 00:44:37,010
process it and the software that enables you to configure those tools and make

341
00:44:37,010 --> 00:44:45,000
them scale so that you can go from a few at once to tens of thousands at once.

342
00:44:45,130 --> 00:44:56,390
And so if you want to run a million of these sequences on the tools that are available most places in the world now to get the answers,

343
00:44:56,930 --> 00:45:01,250
it might take you a month to run them.

344
00:45:02,750 --> 00:45:09,170
The ambition is to have it so you could run them in a day. I don't think they will.

345
00:45:10,190 --> 00:45:14,900
But less than a month away, much less than a.

346
00:45:17,260 --> 00:45:23,589
To me, that's sort of voodoo stuff because it means highly, highly, highly skilled people,

347
00:45:23,590 --> 00:45:29,770
of which there are not many that bring that together to make it perform in that sort of way.

348
00:45:29,810 --> 00:45:36,370
And so that's that's the difference. I don't I don't even get you more massive insights.

349
00:45:36,370 --> 00:45:39,550
You get more rapid turnaround. That would be incremental.

350
00:45:39,650 --> 00:45:48,190
Yeah. Yeah. So, I mean, another project of yours, which was, I guess, again, to do with speeding things up is the land pool.

351
00:45:49,240 --> 00:46:02,080
So did that come about? This was well, we work with Oxford Nanopore Company because they were quite very sweet and because.

352
00:46:05,830 --> 00:46:11,380
They've been sort of local originally a local tech start up and it's really brilliant.

353
00:46:13,180 --> 00:46:17,350
But they are annoying because they're so brilliant they can do all of this stuff.

354
00:46:17,830 --> 00:46:28,920
But they've they've been being quite reasonably, not that attentive to detail things.

355
00:46:29,260 --> 00:46:35,750
And of course, you tend to the new detail. They're getting much better at that, becoming a really professional outfit.

356
00:46:35,800 --> 00:46:40,460
And and they have got to take the group.

357
00:46:42,170 --> 00:46:50,370
Is more than likely to become the winning ticket at the moment to be it.

358
00:46:51,280 --> 00:46:56,530
So let's just go back to the beginning again. We have to bear in mind listeners who might not know the background to this.

359
00:46:56,610 --> 00:47:08,510
What what does that technology do? These are the technologies that take the nucleic acids or the DNA,

360
00:47:08,810 --> 00:47:25,100
RNA of an organism and then mix and process that material to generate an output, which is the sequence, which is the genetic code.

361
00:47:26,000 --> 00:47:33,020
And there are lots of ways of doing it at different levels with different machines.

362
00:47:33,020 --> 00:47:37,849
You can do it on each one of those different sheets. That's it.

363
00:47:37,850 --> 00:47:46,260
So in a very different way. And out of it would come the genetic code.

364
00:47:46,280 --> 00:47:55,280
So it's like if you like listening to music, you could have a radio produce, you could have a tape recorder doing it, you could have a turntable disc.

365
00:47:55,610 --> 00:47:59,350
And they've all got to have their own way of getting music.

366
00:48:00,020 --> 00:48:07,580
And it's a sort of range of devices that sequence that are quite different, that give you a sequence.

367
00:48:08,630 --> 00:48:17,480
And they they constrained by various limitations, each one of them,

368
00:48:18,290 --> 00:48:26,180
that they could slow to make it difficult to run or expensive to run exactly the same audio equipment, in a way.

369
00:48:27,020 --> 00:48:31,459
And Illumina is the most successful one.

370
00:48:31,460 --> 00:48:45,380
So what you might say is that they were like the Sony Walkman dominated the world and other little might come along and leapfrog over them,

371
00:48:45,380 --> 00:48:48,620
which is Apple. Well, in a funny way.

372
00:48:49,280 --> 00:48:56,120
Oxford, Nanopore is a bit like Apple versus Sony Walkman, which still dominates the world at the moment.

373
00:48:56,480 --> 00:49:09,590
I think that it's quite possible that that Apple will leapfrog because they've got some really huge advantages in terms of the speed,

374
00:49:09,740 --> 00:49:18,950
the terms of the cost takes to do it in this sense that it it's it's tiny, tiny miniaturised.

375
00:49:18,950 --> 00:49:25,700
It's small, it's fast, it's cheap. And what has been your role in the collaboration?

376
00:49:26,600 --> 00:49:33,549
Oh, we just worked with them a great deal and published a lot of work on that, of course.

377
00:49:33,550 --> 00:49:37,170
So when they came at that point, they said, look, we would be interested.

378
00:49:37,410 --> 00:49:40,490
Were it's a really neat technology.

379
00:49:41,210 --> 00:49:49,400
So you're testing it again to stop doing it. The government was invested in it and withdrew the investment and no idea why.

380
00:49:51,350 --> 00:50:05,210
It's a story yet to be told possibly. So it was a very, very, very smart way of doing testing that would be no better than a PCR.

381
00:50:06,200 --> 00:50:17,300
It would be faster than doing PCR, but just and it would integrate into into computer infrastructure,

382
00:50:17,780 --> 00:50:27,500
probably better than PCR is actually, funnily enough, a waste of the sequencing technology.

383
00:50:27,680 --> 00:50:31,700
So in some ways, as elegant as that solution was.

384
00:50:34,000 --> 00:50:44,709
The more elegant things that nanopore can can do by but gathering sequence in a more complete and comprehensive

385
00:50:44,710 --> 00:50:59,350
way to serve your lessons with lost San Francisco so is so the the land pool was great we did it very well.

386
00:50:59,350 --> 00:51:08,020
We worked with Sheffield doing it and got regulatory approval, but it would never roll it out.

387
00:51:09,860 --> 00:51:14,180
Are you still using it in your lap? No. You wouldn't do that by using the machine.

388
00:51:14,210 --> 00:51:17,030
The machine sequencing machines are still dead. We still use them.

389
00:51:17,040 --> 00:51:26,990
Yeah, we have just about for the UK Health Security Agency in our routine laboratory research.

390
00:51:27,320 --> 00:51:32,990
We help you with the routine that set up SARS-CoV-2 sequencing.

391
00:51:34,040 --> 00:51:42,629
So we setting it up to get trained in a standard way in the routine lab and that will process through this article.

392
00:51:42,630 --> 00:51:50,110
Related. Cloud peace. And that's an experiment to establish.

393
00:51:52,120 --> 00:51:56,080
Nanopore sequencing cloud results to.

394
00:51:59,320 --> 00:52:02,979
Hopefully a pretty straightforward, seamless way.

395
00:52:02,980 --> 00:52:09,190
So it comes off the sequence that goes through the cloud in about 15, 20 minutes.

396
00:52:09,880 --> 00:52:14,140
Currently, it takes about 4 to 5 hours to run this stuff through.

397
00:52:14,980 --> 00:52:25,690
And this is all got to do with having the compute and being able to parallel the witness data that you're running through that well,

398
00:52:25,690 --> 00:52:29,380
it probably will do it good. Start with smaller scale stuff that's here.

399
00:52:29,740 --> 00:52:39,190
It's likely that the Office of National Statistics samples that have the virus in it will get all sequenced,

400
00:52:39,820 --> 00:52:45,670
and we will probably do that at the moment studying Northumbria, and they do it very well.

401
00:52:47,860 --> 00:53:01,330
Um, but there'll be, you know, ongoing discussions where the Northumbria do focus on other stuff and whether we should do these research related ones.

402
00:53:02,090 --> 00:53:06,730
Yeah, I'm, I think we're pretty neutral for what needs to be done from.

403
00:53:08,970 --> 00:53:09,840
For Battlefield.