1
00:00:05,750 --> 00:00:08,458
Welcome to the Proteomics in Proximity podcast,

2
00:00:08,625 --> 00:00:12,958
where your co-hosts Cindy Lawley and 
Sarantis Chlamydas from Oink proteomics.

3
00:00:13,291 --> 00:00:17,750
Talk about the intersection of proteomics
with genomics for drug target discovery,

4
00:00:17,750 --> 00:00:20,958
the application of proteomics
to reveal disease biomarkers,

5
00:00:21,250 --> 00:00:25,916
and current trends in using proteomics
to unlock biological mechanisms.

6
00:00:26,250 --> 00:00:29,250
Here we have your host,
Cindy, and Sarantis.

7
00:00:29,625 --> 00:00:30,500
Hey there.

8
00:00:30,500 --> 00:00:34,958
Welcome to Proteomics in Proximity
where Sarantis, and I will be talking to

9
00:00:35,291 --> 00:00:36,291
Cornelia today.

10
00:00:36,291 --> 00:00:38,166
I'll have Sarantis
introduce her in a moment.

11
00:00:38,166 --> 00:00:41,666
But first I wanted to announce
a very exciting advance,

12
00:00:41,666 --> 00:00:46,541
in Olink where we have now merged
with Thermo Fisher Scientific.

13
00:00:46,541 --> 00:00:51,375
So we're part of the Proteomics Services
division within, Thermo Fisher.

14
00:00:51,375 --> 00:00:54,458
And we're definitely going 
to be talking about

15
00:00:54,458 --> 00:00:59,750
the ability to sort of sequence
the proteome as well as genotype

16
00:00:59,750 --> 00:01:03,791
the proteome in future episodes,
because these technologies are

17
00:01:03,791 --> 00:01:09,333
incredibly complementary
under this umbrella of this exciting

18
00:01:09,333 --> 00:01:12,625
Thermo Fisher Scientific parent company.

19
00:01:13,250 --> 00:01:17,958
And with that, I'm going to allow Sarantis
to introduce our guest for today.

20
00:01:17,958 --> 00:01:19,875
We're super excited to have Cornelia here.

21
00:01:19,875 --> 00:01:20,916
Sarantis, please.

22
00:01:20,916 --> 00:01:22,750
Thank you very much, Cindy for the introduction.

23
00:01:22,750 --> 00:01:23,708
Thank you very much, Cornelia

24
00:01:23,708 --> 00:01:26,250
for coming with us, it’s
the last episode before summer holidays.

25
00:01:26,250 --> 00:01:30,041
We are really excited to have with us,
Professor Cornelia van Duijn.

26
00:01:30,250 --> 00:01:35,000
She's a professor of epidemiology in the 
Population Health Department of Oxford University.

27
00:01:35,000 --> 00:01:37,750
And today, we’re
going to talk about your exciting work

28
00:01:37,750 --> 00:01:41,666
and mainly dedicated to aging 
and age-related diseases.

29
00:01:42,083 --> 00:01:42,875
Cornelia,

30
00:01:42,875 --> 00:01:43,625
would you like to start

31
00:01:43,625 --> 00:01:45,375
telling us a little bit
about your background

32
00:01:45,375 --> 00:01:47,708
and your scientific interest
and expertise?

33
00:01:47,708 --> 00:01:50,208
Thank you very much for joining our group.

34
00:01:50,208 --> 00:01:52,000
Sure it's my pleasure to be here.

35
00:01:52,000 --> 00:01:53,500
It's the great pleasure.

36
00:01:53,500 --> 00:01:56,375
But, yeah, my background, I think.

37
00:01:56,375 --> 00:02:01,291
I work in epidemiology again,
I studied there, there's an epidemiologist

38
00:02:01,291 --> 00:02:04,500
now one of 30 years ago
working on dementia,

39
00:02:05,083 --> 00:02:08,000
which was in that time
still a forgotten epidemic.

40
00:02:08,000 --> 00:02:11,541
I think everybody swarmed with dementia

41
00:02:11,541 --> 00:02:14,833
in their families, I guess,
particularly parents and grandparents.

42
00:02:14,833 --> 00:02:19,541
But in those days, people 
hadn't heard of the disease, hardly.

43
00:02:19,541 --> 00:02:21,500
Definitely not of Alzheimer's disease

44
00:02:21,500 --> 00:02:25,583
and had difficulty
grabbing what Parkinson's disease is.

45
00:02:25,583 --> 00:02:29,125
But, 
I started out doing the epidemiology,

46
00:02:29,125 --> 00:02:33,958
but then figured that out
pretty soon, that the only risk factor

47
00:02:33,958 --> 00:02:38,083
that we could find in
those days was just family history.

48
00:02:38,083 --> 00:02:41,750
So I switched to genetics. And for long.

49
00:02:41,750 --> 00:02:45,708
I did my PhD
just waiting for the good old markers,

50
00:02:45,708 --> 00:02:49,458
the genetic markers,
to do the linkage analysis in the family.

51
00:02:49,458 --> 00:02:50,500
It's finding the genes.

52
00:02:50,500 --> 00:02:55,625
So this was waiting for months
for a six RFLPs to arrive.

53
00:02:55,625 --> 00:03:01,750
And then two had failed and I went into
another cycle of waiting and waiting.

54
00:03:02,250 --> 00:03:07,375
So in the end of the day we found genes,
and in the end of the day, I was more

55
00:03:07,375 --> 00:03:11,458
than happy that at the technology
emerged to do larger based studies.

56
00:03:11,541 --> 00:03:17,083
And then I went into the genetic 
associations studies genome wide.

57
00:03:17,083 --> 00:03:20,666
And that was millions of millions

58
00:03:20,666 --> 00:03:23,833
of genetic variants
to study in millions of people

59
00:03:24,583 --> 00:03:29,291
and now finally arrived for metabolomics
for the age of proteomics.

60
00:03:29,291 --> 00:03:33,208
So that's the background
and back to epidemiology, not anymore in

61
00:03:33,208 --> 00:03:35,291
Rotterdam but now in Oxford.

62
00:03:35,583 --> 00:03:39,708
So in epidemiology, I think about this
as a challenging field

63
00:03:39,708 --> 00:03:43,625
because often
you're dealing with population level data

64
00:03:43,625 --> 00:03:48,500
sets of community data
that are imperfect, that are messy,

65
00:03:48,500 --> 00:03:52,458
that aren't as clean as I imagine

66
00:03:52,458 --> 00:03:56,500
some of the genetic data sets enable.

67
00:03:56,750 --> 00:04:01,875
Is that a factor in how you've
how you've evolved your career in bringing

68
00:04:01,875 --> 00:04:04,875
in these omics that now you have something

69
00:04:04,875 --> 00:04:07,875
to associate with that maybe is more

70
00:04:08,083 --> 00:04:11,958
I don't know, it's just really hard
to collect environmental data, right?

71
00:04:12,208 --> 00:04:14,833
And epidemiology is plagued with this.

72
00:04:14,833 --> 00:04:17,708
Well, I totally agree with you because I think

73
00:04:17,708 --> 00:04:22,166
if you look at epidemiology
and I'm not only the data analyst,

74
00:04:22,166 --> 00:04:25,958
I also have been able to set up

75
00:04:25,958 --> 00:04:28,958
the different epidemiological studies

76
00:04:29,416 --> 00:04:32,250
and, one of them was a Rotterdam study,

77
00:04:32,250 --> 00:04:35,625
with the
elderly people really followed over time.

78
00:04:36,125 --> 00:04:38,875
And it's hard. It's a lot of effort.

79
00:04:38,875 --> 00:04:44,375
And sometimes I wonder that people,
young people

80
00:04:44,375 --> 00:04:48,333
who are dealing with all these data now
think why haven't they done it better?

81
00:04:48,333 --> 00:04:50,500
But it's a huge effort.

82
00:04:50,500 --> 00:04:53,875
Not only the Rotterdam study,
we set up large family

83
00:04:53,875 --> 00:04:56,041
based studies, like the Erasmus Rettfeld study

84
00:04:56,041 --> 00:04:59,375
and last but not least Generation R

85
00:05:00,208 --> 00:05:03,708
Wasn’t the leading in there,
but I was working on there, setting up

86
00:05:03,708 --> 00:05:09,791
and a study of, little children,
followed from utero.

87
00:05:09,791 --> 00:05:12,000
And it is hard.

88
00:05:12,000 --> 00:05:16,166
It's hard to get really a grasp
on how you capture,

89
00:05:16,250 --> 00:05:19,208
to what people are exposed to.

90
00:05:19,208 --> 00:05:22,166
And then, of course,
if you think about people,

91
00:05:22,166 --> 00:05:28,333
the exposures that you have over
time are changing, they’re ever changing.

92
00:05:28,833 --> 00:05:32,125
Your smoking habits,
your alcohol habits.

93
00:05:32,125 --> 00:05:37,708
What your weight is and what you're
eating, incredibly changes over time.

94
00:05:37,708 --> 00:05:42,583
And, I think it's the availability,
the cost, but also definitely,

95
00:05:43,166 --> 00:05:45,958
what you know is healthy and unhealthy.

96
00:05:45,958 --> 00:05:51,666
So, they're growing inside,
but what we’re learning to know now

97
00:05:51,708 --> 00:05:55,791
that it's important to do these studies.

98
00:05:55,791 --> 00:06:03,916
And they have been incredibly helpful
in making the genetics study happen.

99
00:06:03,916 --> 00:06:08,000
It has enabled it
that it would have not been at the states

100
00:06:08,000 --> 00:06:11,166
where it is now without it, but definitely

101
00:06:11,166 --> 00:06:15,000
also, 
a lot of the future of proteomics will be

102
00:06:15,000 --> 00:06:18,500
in these studies
so we’re depending on them.

103
00:06:19,458 --> 00:06:21,125
That's a great.

104
00:06:21,125 --> 00:06:22,958
No, go a head, Sarantis.

105
00:06:22,958 --> 00:06:23,750
I just wanted to follow up

106
00:06:23,750 --> 00:06:26,958
on this question that you have posed,
know, for
genetics and proteomics.

107
00:06:27,583 --> 00:06:30,583
Nowadays, for example, these complicated
diseases like Alzheimer's,

108
00:06:31,083 --> 00:06:34,541
do you think one omic is enough
and how you see multi omics in this field?

109
00:06:34,833 --> 00:06:38,291
How you see the challenge that people
that are facing of data integration.

110
00:06:38,291 --> 00:06:39,541
What is your feeling on this?

111
00:06:39,541 --> 00:06:44,833
Well, I think, we learn a lot from
genetics, and I think you can't deny that.

112
00:06:44,833 --> 00:06:50,666
So people have had troubles with it
that once you start doing at scale

113
00:06:50,666 --> 00:06:54,208
as genome wide
association studies, 

114
00:06:54,875 --> 00:06:57,875
we’re just going to the moon,

115
00:06:58,166 --> 00:07:01,000
And then beyond,
we were almost going to Mars,

116
00:07:01,000 --> 00:07:05,375
just finding new pathways
in the disease process.

117
00:07:05,375 --> 00:07:06,958
And then of course, people said, 

118
00:07:06,958 --> 00:07:11,375
“well, we knew this, we always get this.”
But we finally have it established.

119
00:07:11,375 --> 00:07:13,041
And that is what you do with genomics.

120
00:07:13,041 --> 00:07:16,708
I mean, you can hypothesize
that the complement,

121
00:07:16,791 --> 00:07:20,041
system as one of the immune systems,

122
00:07:20,041 --> 00:07:24,208
that is one of the defenses against that

123
00:07:24,208 --> 00:07:29,291
invasive swarm of bacteria and viruses
that you can have the hypothesis.

124
00:07:29,291 --> 00:07:33,041
And it was there already before,
the theory was, that it’s implicated

125
00:07:33,041 --> 00:07:36,708
in your pathogenesis,
the development of dementia.

126
00:07:37,333 --> 00:07:40,000
But, you know, genetics nailed it.

127
00:07:40,000 --> 00:07:41,208
It benchmarked it.

128
00:07:41,208 --> 00:07:42,041
It says, well,

129
00:07:42,041 --> 00:07:45,708
if we have genes, I'm not sufficient
there, you’re not going to make it.

130
00:07:46,333 --> 00:07:51,541
In genetics of dementia,
we went through the whole series of

131
00:07:51,541 --> 00:07:54,875
we though it's a neuronal disease
because your neurons don’t

132
00:07:54,916 --> 00:07:56,916
function anymore
and therefore you're demented.

133
00:07:56,916 --> 00:07:59,291
You forget things,
you can’t even comprehend things,

134
00:07:59,291 --> 00:08:02,458
how to put your shoes on
and where you should put them.

135
00:08:02,958 --> 00:08:04,791
You put them on your head.

136
00:08:04,791 --> 00:08:08,166
All the things for your brain to work,

137
00:08:08,541 --> 00:08:12,625
of course it's the neurons that die
and that, give you the disease

138
00:08:12,625 --> 00:08:16,750
that will make you forget things
and not understand things.

139
00:08:17,291 --> 00:08:20,125
But then in the end of the day,
what we learned from GWAS

140
00:08:20,125 --> 00:08:24,708
is that the microglia, the helper cells
of your neurons, were much more important.

141
00:08:25,208 --> 00:08:30,000
So definitely we learned a lot of it,
what we did not learn.

142
00:08:30,000 --> 00:08:30,708
And that's always

143
00:08:30,708 --> 00:08:34,875
as the scientist, for young scientist,
that's even more important, right?

144
00:08:34,875 --> 00:08:39,500
It wasn't the endpoint because what
we learned from genetics, for instance,

145
00:08:39,500 --> 00:08:43,833
that the apolipoprotein E4 variant

146
00:08:43,833 --> 00:08:47,333
more or less splits
the population in half,

147
00:08:47,333 --> 00:08:51,250
who gets the disease and determines
who gets the disease early or late.

148
00:08:51,833 --> 00:08:56,541
But you know, it doesn't tell you
whether you get it that 16, 17 or 18.

149
00:08:57,416 --> 00:09:02,541
That is so important for people
and for that, you need these proteins

150
00:09:02,750 --> 00:09:05,625
or the metabolites, that will tell you.

151
00:09:05,625 --> 00:09:09,583
And that's what we see
now that B tau is telling you that.

152
00:09:09,916 --> 00:09:13,625
But we see also, that other proteins

153
00:09:13,625 --> 00:09:16,666
like GFAP and that NFL

154
00:09:18,041 --> 00:09:20,708
that you can measure
easily that there's also doing that.

155
00:09:20,708 --> 00:09:25,250
And that is incredibly important
and that is what we need to know.

156
00:09:25,250 --> 00:09:28,000
And that is what we need to take further.

157
00:09:28,000 --> 00:09:31,375
So I’ll ask a question now 
along that genetics line.

158
00:09:31,375 --> 00:09:36,125
Along with Rotterdam study,
Generation R, certainly CHARGE initiatives.

159
00:09:36,125 --> 00:09:38,416
And all the cohorts
that are involved in that.

160
00:09:38,416 --> 00:09:42,333
You have been involved
in a lot of really pivotal work

161
00:09:42,333 --> 00:09:44,625
in that population health area.

162
00:09:44,625 --> 00:09:48,166
One of the other Biobanks
I've seen you involved with

163
00:09:48,166 --> 00:09:50,416
is the China Kadoorie Biobank.

164
00:09:50,416 --> 00:09:53,916
That's incredibly important
for our understanding

165
00:09:53,916 --> 00:09:57,083
of East Asian populations
and how they're very different from

166
00:09:57,083 --> 00:10:00,416
what we see in the UK Biobank
as just another example.

167
00:10:00,875 --> 00:10:03,250
And I just saw in Oxford

168
00:10:03,250 --> 00:10:06,250
a presentation given by,
I believe it was Alfred,

169
00:10:06,291 --> 00:10:09,500
who talked about, GWAS

170
00:10:09,791 --> 00:10:12,875
leveraging proteomics
in the context of genomics

171
00:10:12,875 --> 00:10:16,541
with the clinical data that are
that are available for these cohorts.

172
00:10:16,541 --> 00:10:22,916
Can you talk a little bit about the outliers
and liars that we talked about there?

173
00:10:23,083 --> 00:10:29,291
And just explain how proteins are 
showing signals about lifestyle factors

174
00:10:29,291 --> 00:10:31,750
that I think is pretty compelling.

175
00:10:33,041 --> 00:10:38,375
Yeah, sure I think what has been
a breakthrough in that, not with my head

176
00:10:38,375 --> 00:10:42,125
as a geneticist, but with the other head
as an epidemiologist,

177
00:10:42,125 --> 00:10:45,791
because after all, I'm a genetic 
epidemiologist by training.

178
00:10:46,333 --> 00:10:50,375
Is that what the proteomics is giving us

179
00:10:50,375 --> 00:10:55,625
is really the mirror of what happens
if you have an exposure that is,

180
00:10:56,750 --> 00:10:59,833
in the case of smoking,
I think nobody doubts anymore

181
00:10:59,833 --> 00:11:04,916
that that is shortening your lifespan,
is giving you increased risk of cancer,

182
00:11:05,000 --> 00:11:07,833
but also lung diseases,
cardiovascular diseases,

183
00:11:07,833 --> 00:11:12,625
and definitely in the end of the time
also it's related to

184
00:11:12,625 --> 00:11:16,291
many a neurological diseases

185
00:11:16,291 --> 00:11:19,625
and neurodegenerative 
diseases like dementia.

186
00:11:19,625 --> 00:11:23,708
But measuring
these exposures is a nightmare.

187
00:11:23,708 --> 00:11:25,875
And it's difficult for smoking.

188
00:11:25,875 --> 00:11:29,583
And there’s people specialized
in how to asses how much you smoke.

189
00:11:29,583 --> 00:11:33,958
But it's quite a difficult task.

190
00:11:33,958 --> 00:11:36,250
So you have to ask,
when did you start smoking?

191
00:11:36,250 --> 00:11:37,625
When did you stop smoking?

192
00:11:37,625 --> 00:11:39,875
How much the smoke over time.

193
00:11:39,875 --> 00:11:44,875
Because that everybody thinks, oh,
I smoke half a packets or a packet today.

194
00:11:44,875 --> 00:11:47,666
I only have smoke today, 24 cigarettes.

195
00:11:47,666 --> 00:11:49,833
I do have to take another one, would I?

196
00:11:49,833 --> 00:11:51,916
So it's approximation.

197
00:11:51,916 --> 00:11:53,500
Nobody will live like that.

198
00:11:53,500 --> 00:11:57,875
People stop smoking when they're pregnant
or the first child is born.

199
00:11:57,875 --> 00:12:00,416
You think I'd have to be more healthy now?

200
00:12:00,416 --> 00:12:02,416
It's quite an effort.

201
00:12:02,416 --> 00:12:05,541
And don't get us
started as epidemiologists

202
00:12:05,541 --> 00:12:09,375
on something more complex like,
alcohol use.

203
00:12:09,375 --> 00:12:12,583
Because alcohol use, we have the month

204
00:12:12,583 --> 00:12:17,166
that we're all asked to be sober October
or dry January.

205
00:12:17,166 --> 00:12:20,166
And that becomes even more difficult.

206
00:12:20,208 --> 00:12:22,416
Definitely there is the pregnancy issue.

207
00:12:22,416 --> 00:12:26,333
Definitely there is,
once you start being older,

208
00:12:26,333 --> 00:12:29,333
you can't deal with it
anymore as well as before.

209
00:12:29,750 --> 00:12:32,083
So what do we do now?

210
00:12:32,083 --> 00:12:35,666
Well,
we really ventured out targeted smoking

211
00:12:35,666 --> 00:12:40,416
because it is the major determinant
of your life expectancy

212
00:12:40,416 --> 00:12:43,875
and all the diseases that you'll 
encounter with the old age.

213
00:12:43,875 --> 00:12:47,083
So the question was, what is really

214
00:12:47,083 --> 00:12:50,833
the proteomic profile
associated with smoking?

215
00:12:50,833 --> 00:12:54,500
And see how [---]

216
00:12:54,666 --> 00:12:58,625
really ventured out on this 
he had an interesting cancer.

217
00:12:58,625 --> 00:13:02,875
And of course lung cancer,
very well known as the major outcome.

218
00:13:03,541 --> 00:13:07,708
And what we did
see in the very simple experiment,

219
00:13:07,708 --> 00:13:11,625
seeing whether we could discriminate
those who were never smokers

220
00:13:11,625 --> 00:13:13,333
or told us that were never smokers,

221
00:13:13,333 --> 00:13:17,208
and those who were currently smoking
and had been honest about that.

222
00:13:17,791 --> 00:13:22,333
We saw that we would set the data to,
I mean, really quite well

223
00:13:22,333 --> 00:13:25,833
or using the proteomics,
and then you really talking about

224
00:13:25,833 --> 00:13:28,833
the discrimination of 0.95,

225
00:13:29,333 --> 00:13:33,041
you hardly see that in any 
epidemiological setting.

226
00:13:33,750 --> 00:13:35,083
Well, that was fantastic.

227
00:13:35,083 --> 00:13:37,916
But we still saw overlap
between the two groups.

228
00:13:37,916 --> 00:13:40,833
And I know that is the major question.

229
00:13:40,833 --> 00:13:44,458
So if you are a never smoker,
you declare yourself as a never smoker,

230
00:13:44,458 --> 00:13:48,708
and then 
you still have a proteome profile

231
00:13:49,791 --> 00:13:50,708
that looks like

232
00:13:50,708 --> 00:13:53,708
you are quite a heavy smoker.

233
00:13:54,541 --> 00:13:56,833
It raises questions
and that is the fantastic thing.

234
00:13:56,833 --> 00:14:01,500
So we thought, if may be that these 
people have not been fully honest

235
00:14:01,500 --> 00:14:04,541
or they forgot 
that they ever smoked

236
00:14:05,083 --> 00:14:08,708
or they didn't want to be reminded
of the fact that they ever smoked.

237
00:14:09,208 --> 00:14:11,291
And that is certainly the case.

238
00:14:11,291 --> 00:14:16,625
And we noticed for instance of alcohol
that people say, I’m not drinking alcohol.

239
00:14:16,625 --> 00:14:21,833
And they turned out to be ex users
that have to stop because some problem

240
00:14:22,416 --> 00:14:25,416
that was related to alcohol,
for instance, the liver.

241
00:14:26,166 --> 00:14:29,666
But there's also
alternative explanations.

242
00:14:29,666 --> 00:14:34,333
And that was the important thing
that, we really soon found out

243
00:14:34,333 --> 00:14:38,708
that if you look at this profile,
it's really determined

244
00:14:39,458 --> 00:14:42,833
at least half of it
in the general population by smoker.

245
00:14:42,833 --> 00:14:47,375
I used smoker that determines
how high your score is in

246
00:14:47,375 --> 00:14:52,375
what we call P -SIN, how much you've seen
in terms of your smoking habits.

247
00:14:52,875 --> 00:14:58,208
But, 
if you really, look at to other factors

248
00:14:58,208 --> 00:15:00,833
that may determine this score,

249
00:15:00,833 --> 00:15:03,791
how can it be
if we talked to a genetic epidemiologist,

250
00:15:03,791 --> 00:15:07,833
we looked at the genes and there's some
contribution of the genes but not big.

251
00:15:09,208 --> 00:15:12,166
If you'll look at the exposures.

252
00:15:12,166 --> 00:15:14,416
Well see all the exposures.

253
00:15:14,416 --> 00:15:19,208
So one of the most fantastic thing is that
we found that your maternal smoker,

254
00:15:19,208 --> 00:15:23,208
whether you're not a smoker,
was popping up.

255
00:15:23,791 --> 00:15:27,041
Whether you were passively smoking,
popped up.

256
00:15:27,375 --> 00:15:30,791
How much air pollution was 
around you, popped up.

257
00:15:31,375 --> 00:15:34,083
But there's also all these factors

258
00:15:34,083 --> 00:15:37,083
that we thought, hey,
also obesity pops up.

259
00:15:37,083 --> 00:15:40,333
And if you know a little bit
about smoking, it's

260
00:15:41,458 --> 00:15:44,291
it’s one of the strange things 
is if you smoke you'll

261
00:15:44,291 --> 00:15:47,291
usually have a lower weight
than nonsmokers.

262
00:15:47,458 --> 00:15:52,916
If you stop smoking, a lot of people
say I go obese and I don't want that,

263
00:15:52,916 --> 00:15:57,791
I don't fit in my dress anymore, and
I don't look as beautiful as I did before.

264
00:15:57,791 --> 00:15:59,583
So that is affected.

265
00:15:59,583 --> 00:16:01,541
That did not surprise us.

266
00:16:01,541 --> 00:16:05,541
And if think about how to explain this,
we also started seeing that

267
00:16:05,541 --> 00:16:10,625
there are probably common pathways which
go to aging and age related diseases.

268
00:16:10,625 --> 00:16:15,375
With overlap
for instance for obesity and smoking.

269
00:16:15,916 --> 00:16:19,541
That is really what you expect
also because and we don't think

270
00:16:20,083 --> 00:16:23,666
that smoking has a unique pathway.

271
00:16:23,666 --> 00:16:25,000
It may be in your lungs,

272
00:16:26,291 --> 00:16:30,125
 I mean, in direct exposure.

273
00:16:30,125 --> 00:16:32,500
The oesophagus also, right.

274
00:16:32,500 --> 00:16:35,000
We all know, that is a problem.

275
00:16:35,000 --> 00:16:39,791
But really if you start thinking
how it causes aging, of course,

276
00:16:40,000 --> 00:16:43,500
we all know
that if you ask your pathologist, well,

277
00:16:43,500 --> 00:16:48,208
you will not ask your own pathologist,
but that of a another person.

278
00:16:49,125 --> 00:16:52,125
And if you look at the skin,

279
00:16:52,750 --> 00:16:55,750
really,
if you look at the in the microscope,

280
00:16:55,916 --> 00:16:59,041
you really see something
awkward in the smokers.

281
00:16:59,041 --> 00:17:03,958
The skin ages and we all see that
your throat, you're voice.

282
00:17:03,958 --> 00:17:08,291
Usually, people who are 80 years and
have smoked all their life,

283
00:17:09,041 --> 00:17:11,833
you hear, oh, this is a course voice.

284
00:17:11,833 --> 00:17:14,500
So we do see differences.

285
00:17:14,500 --> 00:17:18,208
But the processes
that are ongoing in your body overlap.

286
00:17:18,375 --> 00:17:22,833
So we also saw that of course
we think some people don't tell us anymore

287
00:17:22,833 --> 00:17:25,500
whether they smoke.
And how much they smoke.

288
00:17:25,500 --> 00:17:29,625
But, we also
think that there are other reasons.

289
00:17:29,791 --> 00:17:34,291
But some of the reasons are, you know,
we can't put our finger on it.

290
00:17:34,291 --> 00:17:38,916
But the other common ones, like obesity,
it's the major problem worldwide, so

291
00:17:40,083 --> 00:17:41,416
we see it.

292
00:17:41,750 --> 00:17:44,083
I’ll also correct myself.

293
00:17:44,083 --> 00:17:45,166
It wasn’t Alfred.

294
00:17:45,166 --> 00:17:49,458
Alfred talked about GWAS in the 
China Kadoorie Biobank,

295
00:17:49,458 --> 00:17:52,375
but it was Sihao that actually
presented this.

296
00:17:52,375 --> 00:17:54,500
Sihao is a PhD student

297
00:17:54,500 --> 00:17:58,666
who has been working with these data
and looking in the UK B data as well

298
00:17:58,666 --> 00:18:03,125
as corroborating in China Kadoorie
Biobank, B data, super, super interesting.

299
00:18:03,125 --> 00:18:07,250
So that that piece and that this idea
of having a smoking signature

300
00:18:07,500 --> 00:18:10,958
and an ability to determine
and maybe it's, you know,

301
00:18:10,958 --> 00:18:14,791
secondhand smoking and heavy secondhand
smoking or something like that.

302
00:18:14,791 --> 00:18:19,416
But I think being able to parse this out
and corroborate the genetics

303
00:18:19,791 --> 00:18:22,541
and the proteomics in any way with,

304
00:18:22,541 --> 00:18:27,041
the epidemiological data
and vice versa is super exciting.

305
00:18:27,041 --> 00:18:30,041
And then, of course, we've talked
on this podcast before about using

306
00:18:30,041 --> 00:18:33,500
genetics to corroborate proteomics
and proteomics to corroborate,

307
00:18:33,791 --> 00:18:37,250
what we're seeing in the, in the genetics
that have maybe supported

308
00:18:37,250 --> 00:18:39,208
drug programs, for example.

309
00:18:39,208 --> 00:18:43,875
So can we, and this is 
Sarantis’ absolute area of expertise,

310
00:18:43,875 --> 00:18:46,750
if we could transition to aging,

311
00:18:46,750 --> 00:18:48,041
That's a great point, actually.

312
00:18:48,041 --> 00:18:51,041
You know, I’m intriguing for the fact
we say the mothers when they are pregnant

313
00:18:51,041 --> 00:18:53,291
and they're smoking,
you see effects on the babies.

314
00:18:53,291 --> 00:18:54,791
There are a lot of studies like that.

315
00:18:54,791 --> 00:18:57,083
That means apart genetics,
there are a lot of other factors,

316
00:18:57,083 --> 00:19:00,958
probably epigenetics
that may influence all of this transition.

317
00:19:01,333 --> 00:19:04,083
And we know for measuring the aging

318
00:19:04,083 --> 00:19:07,000
epigenetic clocks are really
the gold standard so far.

319
00:19:07,000 --> 00:19:10,000
But proteomics takes a really big

320
00:19:10,958 --> 00:19:14,125
attention and really go to nail

321
00:19:14,125 --> 00:19:17,958
down the details of aging
and aging related disease.

322
00:19:17,958 --> 00:19:18,333
Right.

323
00:19:18,333 --> 00:19:19,500
And you have seen these

324
00:19:19,500 --> 00:19:23,625
with your own data and with amazing work
we had with Austin together.

325
00:19:23,625 --> 00:19:27,875
And It would be soon published.

326
00:19:27,875 --> 00:19:30,291
Would you like to say a few words
about the biological age

327
00:19:30,291 --> 00:19:33,916
and how proteomics clock enable
the study of biological age?

328
00:19:34,708 --> 00:19:35,666
That'd be great.

329
00:19:35,666 --> 00:19:36,041
Yeah.

330
00:19:36,041 --> 00:19:41,166
I think one of the the golden grails
we're all looking for is how to live long

331
00:19:41,166 --> 00:19:45,750
and how to not to become older
looking than you are, right.

332
00:19:45,791 --> 00:19:47,875
And it's a it's a golden grail.

333
00:19:47,875 --> 00:19:52,125
And I think this longevity
research, has been

334
00:19:52,750 --> 00:19:55,166
what has baffled me for

335
00:19:55,166 --> 00:19:58,833
always and I’ve been really 
working on aging 

336
00:19:58,833 --> 00:20:02,166
and dementia now 
already 30 or something more.

337
00:20:02,166 --> 00:20:03,208
That there was a lot of progress

338
00:20:03,208 --> 00:20:06,875
in the field of animal
based experimental studies.

339
00:20:07,666 --> 00:20:10,958
And they had wonderful findings,
whether it was telomeres.

340
00:20:10,958 --> 00:20:14,791
Whether it was on

341
00:20:14,791 --> 00:20:18,666
the basis of 
protein homeostasis or metabolites.

342
00:20:18,708 --> 00:20:21,708
IGF 1 was a notorious one.

343
00:20:21,875 --> 00:20:24,583
And all these things seem to fit, right.

344
00:20:24,583 --> 00:20:27,791
All the animals,
if you look at the animal kingdom

345
00:20:27,791 --> 00:20:32,375
except for the birds, but the smaller
animals live longer than the other animals

346
00:20:33,000 --> 00:20:36,041
and the wonderful study, 

347
00:20:36,041 --> 00:20:41,833
dogs in science with undercover
a big dog life expectancy 6 to 8 years,

348
00:20:41,833 --> 00:20:45,583
if it’s a Danish dog or a big pointer,

349
00:20:45,583 --> 00:20:54,333
a small dog with a very long life
expectancy of 15 years, 20 years.

350
00:20:55,125 --> 00:20:59,375
But it never translated to humans,
and that has bothered me forever.

351
00:20:59,375 --> 00:21:02,666
So even something like telomeres again,
the Nobel Prize, right.

352
00:21:02,708 --> 00:21:07,500
So as a Nobel Prize on it,
it works the most well ever.

353
00:21:07,500 --> 00:21:09,708
And in the animal it works.

354
00:21:09,708 --> 00:21:14,541
Except in humans you do see associations,
you do see suggestions.

355
00:21:14,541 --> 00:21:18,791
You don't see a lot, a lot, a lot,

356
00:21:18,791 --> 00:21:22,958
if you translate it to diseases
has been the breakthrough.

357
00:21:22,958 --> 00:21:25,958
If we look at the proteomics clock now,

358
00:21:26,041 --> 00:21:28,875
and if you look how it

359
00:21:28,875 --> 00:21:32,708
predicts, projects
to diseases, it's phenomenal.

360
00:21:32,708 --> 00:21:37,708
And in that sense if you compare it
with the methylation clock.

361
00:21:38,083 --> 00:21:42,041
Well the first thing I did
you say well whatever we're going to do,

362
00:21:42,041 --> 00:21:46,125
compare first what the overlap is
with the methylation clock.

363
00:21:46,125 --> 00:21:50,375
And I was really understanding that

364
00:21:50,375 --> 00:21:54,666
whatever you find in methylation
also very much goes to this.

365
00:21:55,291 --> 00:21:58,416
I was already up to date

366
00:21:58,416 --> 00:22:03,666
that, you know, a lot on the cancer field
and methylation, huge progress,

367
00:22:03,708 --> 00:22:07,166
 it' seen as a very helpful
and promising field.

368
00:22:08,083 --> 00:22:12,250
But I was actually surprised how 
few evidence there is

369
00:22:12,250 --> 00:22:17,166
for direct links between the
proteomics group and diseases,

370
00:22:17,166 --> 00:22:20,291
and definitely, as with so many diseases

371
00:22:20,291 --> 00:22:23,291
as we see now with the proteomics.

372
00:22:23,291 --> 00:22:26,833
So we were talking a lot
the methylation folks,

373
00:22:26,833 --> 00:22:30,916
and we were just arguing like, okay,
we worked a bit on it,

374
00:22:30,916 --> 00:22:35,250
and definitely [---] worked on it

375
00:22:35,250 --> 00:22:38,625
in relation to psychiatric diseases.

376
00:22:39,041 --> 00:22:42,041
But and we were a little bit amazed
that the

377
00:22:42,041 --> 00:22:46,041
overlap between the proteomics and the
methylation clocks isn't big.

378
00:22:46,041 --> 00:22:49,458
But what you also saw
that in the methylation clocks

379
00:22:49,458 --> 00:22:52,208
what you usually have to tweak that the

380
00:22:52,875 --> 00:22:56,041
the methylation clocks
only associate to disease.

381
00:22:56,041 --> 00:22:59,041
If you are any focusing all

382
00:22:59,041 --> 00:23:04,041
coding proteins at the methylation
that is related

383
00:23:04,041 --> 00:23:07,208
to genes that are known to be involved
in diseases.

384
00:23:08,083 --> 00:23:11,125
It's not so strange
because if you really start thinking

385
00:23:11,125 --> 00:23:14,458
what the what methylation does,
it will be agnostic.

386
00:23:14,458 --> 00:23:18,083
It's just going all over the genome.

387
00:23:18,083 --> 00:23:20,000
The CPT unit.

388
00:23:20,083 --> 00:23:24,458
And what we know of the genome,
only a small fraction

389
00:23:25,041 --> 00:23:28,416
is involved in coding protein.

390
00:23:28,750 --> 00:23:32,208
Now of course
we all think that in a translation

391
00:23:32,208 --> 00:23:36,250
and RNA regulation is important
in the development of the disease.

392
00:23:36,666 --> 00:23:40,166
But in the end of the day,
it's still the protein

393
00:23:40,166 --> 00:23:41,583
who does a lot of the job.

394
00:23:41,583 --> 00:23:43,166
Exactly.

395
00:23:43,958 --> 00:23:47,875
In Alzheimer's
and dementia and vascular dementia,

396
00:23:47,875 --> 00:23:51,750
it's the most important the proteins there.

397
00:23:51,791 --> 00:23:55,583
But I think what we are seeing that
the proteins are also mentioned in

398
00:23:55,625 --> 00:23:56,458
cardiovascular disease.

399
00:23:56,458 --> 00:23:59,208
And it's not unexpected, is it?

400
00:23:59,208 --> 00:24:02,208
It's it's more I expect that the,

401
00:24:02,416 --> 00:24:05,291
the metabolome for instance, did

402
00:24:05,291 --> 00:24:07,000
much less than the proteome.

403
00:24:07,000 --> 00:24:12,125
And that that brings us back to work that
that this is probably the field to be in.

404
00:24:12,958 --> 00:24:16,833
It feels like it's
the druggable aspect of the omics as well.

405
00:24:17,166 --> 00:24:20,291
So the fact that we do have antibody

406
00:24:20,291 --> 00:24:24,791
therapies that are able to target
pathways, I think means that

407
00:24:24,791 --> 00:24:28,625
the translation feels like
it will be more straightforward.

408
00:24:28,958 --> 00:24:32,708
But I think
we're only scratching the surface.

409
00:24:32,708 --> 00:24:37,541
I think well, what I tell any
all young people in my group,

410
00:24:37,750 --> 00:24:43,375
and also others that I come across now,
is that you really has to invest in this.

411
00:24:43,375 --> 00:24:48,250
And I confess to you
and to the world, I always was

412
00:24:48,250 --> 00:24:52,000
a metabolomics fan and I thought
that is going to make it happen.

413
00:24:52,000 --> 00:24:53,666
And that is the place to be

414
00:24:53,666 --> 00:24:57,208
because it's the active compounds, it's
the activated part

415
00:24:57,958 --> 00:25:02,625
and if you compare that now
to the development in proteomics,

416
00:25:02,625 --> 00:25:07,708
I do agree with you, Cindy, it's more
the druggable part in it,

417
00:25:07,708 --> 00:25:11,791
but it's also the part that explains
for us, the thing is, and that makes you

418
00:25:11,791 --> 00:25:13,958
wonder a little bit what's happening.

419
00:25:13,958 --> 00:25:15,541
It's the phenotype, right?

420
00:25:15,541 --> 00:25:17,833
The proteins are really depicting the real phenotype.

421
00:25:17,833 --> 00:25:18,916
Yeah, definitely.

422
00:25:19,250 --> 00:25:24,291
If you go to CPGs, they are, like, more upstream,
like more going to the mechanistic.

423
00:25:24,291 --> 00:25:25,958
That will be other factors
that may influence.

424
00:25:25,958 --> 00:25:27,833
But at the end, end point is the protein.

425
00:25:27,833 --> 00:25:30,833
The real phenotype is what 
happened at the protein level, right?

426
00:25:30,833 --> 00:25:32,125
And that's the real picture.

427
00:25:32,291 --> 00:25:35,500
What worries me also a 
little bit if you are looking at

428
00:25:35,500 --> 00:25:37,458
expression data in the brain, now

429
00:25:37,458 --> 00:25:40,208
and there's often
not a correlation between the two.

430
00:25:40,208 --> 00:25:42,333
And they often go opposite direction.

431
00:25:42,333 --> 00:25:46,500
So that makes us worry a little bit
what's going on there.

432
00:25:46,666 --> 00:25:48,916
I mean you should ask ourselves

433
00:25:48,916 --> 00:25:51,583
what will be the height of the day
in five years.

434
00:25:51,583 --> 00:25:55,458
But the idea now is that,
it's the proteomics

435
00:25:55,458 --> 00:25:57,541
that matters more than anything else.

436
00:25:57,708 --> 00:25:59,083
Exactly.

437
00:25:59,291 --> 00:26:05,291
It's nice to hear that it's adding value 
to the data sets we've got already.

438
00:26:05,458 --> 00:26:06,500
I think there's

439
00:26:07,083 --> 00:26:12,750
the in -depth pathway analysis trying 
to dig into why RNA would go one direction

440
00:26:12,750 --> 00:26:14,541
and proteins would go the other direction.

441
00:26:14,541 --> 00:26:17,583
If we can at least come up with some hypotheses 

442
00:26:17,583 --> 00:26:20,583
for any given system why that would be,

443
00:26:20,583 --> 00:26:23,583
for example, maybe the

444
00:26:23,583 --> 00:26:27,500
products are being cleared out to move to a 
different place where they're being used.

445
00:26:27,500 --> 00:26:30,458
Maybe they're in vesicles or something like that.

446
00:26:30,458 --> 00:26:39,583
Being able to sort of dig in to provide hypotheses 
for testing the mechanism is exciting.

447
00:26:39,583 --> 00:26:45,208
And it means that if people are listening to this podcast 
thinking they wanna go do their PhD,

448
00:26:45,208 --> 00:26:51,541
there are so many questions to answer 
and they should consider going to Oxford,

449
00:26:51,541 --> 00:26:52,333
I will say.

450
00:26:52,333 --> 00:26:53,583
Definitely, definitely.

451
00:26:54,083 --> 00:26:55,291
So I echo that.

452
00:26:55,291 --> 00:27:00,916
I think that, I, I noticed that, 
and it really is the same as genetics.

453
00:27:00,916 --> 00:27:05,916
I mean, we weren't doing the genetic,
the genome wide association study

454
00:27:05,916 --> 00:27:09,041
that we had found 
three genes for diabetes.

455
00:27:09,041 --> 00:27:12,791
And then people said, oh,
we got to find out what these

456
00:27:12,791 --> 00:27:15,166
genes do, and this is probably it.

457
00:27:15,166 --> 00:27:16,791
There's no other genes to be found.

458
00:27:17,083 --> 00:27:20,375
Well, afterwards we found hundreds more.

459
00:27:21,083 --> 00:27:25,958
I mean, that is what we are
at the stage with the proteomics.

460
00:27:25,958 --> 00:27:28,125
I mean, this is the start.

461
00:27:28,125 --> 00:27:31,083
It looks fantastic. It looks great.

462
00:27:31,083 --> 00:27:35,958
But we are at the start,
this will be an effort of 10, 15 years

463
00:27:35,958 --> 00:27:39,083
like it was with genome association studies

464
00:27:39,083 --> 00:27:40,250
We’ve been working on it,

465
00:27:40,250 --> 00:27:41,875
and we still haven't finalized it,

466
00:27:41,875 --> 00:27:47,125
but we have now, genetic risk factors
that we all add together,

467
00:27:47,125 --> 00:27:50,750
the picture is becoming
completely more and more clear.

468
00:27:51,541 --> 00:27:53,333
And in is work in progress.

469
00:27:53,333 --> 00:27:56,291
I mean, we know that from the genetics.

470
00:27:56,291 --> 00:27:59,208
We were staring at the genome
-wide association studies.

471
00:27:59,208 --> 00:28:04,208
We said, oh, we don't see amyloid
at all in working in the genome

472
00:28:04,875 --> 00:28:10,250
Five years later we go into GWAS and
that was the first pathway was amyloid.

473
00:28:10,250 --> 00:28:11,333
The second pathway,

474
00:28:11,333 --> 00:28:13,166
the third one was pathway.

475
00:28:13,166 --> 00:28:15,583
And we asked, what is happening here?

476
00:28:15,583 --> 00:28:18,708
I asked my friend's colleague
and he said, well, we looked at it too,

477
00:28:18,708 --> 00:28:23,083
but what happened is that the people
doing more research in the biochemistry

478
00:28:23,083 --> 00:28:26,291
and start linking those genes to amyloid 
completely.

479
00:28:26,291 --> 00:28:28,291
Now we can go the reverse way.

480
00:28:28,291 --> 00:28:32,708
We we can look at the proteins
associated with the disease.

481
00:28:32,708 --> 00:28:36,500
And of course with now checking
whether they also associate

482
00:28:36,500 --> 00:28:38,416
to the genes of the disease

483
00:28:38,416 --> 00:28:41,416
and the exposures related to the disease.

484
00:28:41,416 --> 00:28:44,416
So it's one of the most exciting tangles,

485
00:28:44,750 --> 00:28:47,583
if you are interested in the disease

486
00:28:47,583 --> 00:28:51,750
and understanding disease,
but also predicting disease,

487
00:28:51,958 --> 00:28:55,875
it's the breaking point,
but don't see it as end points yet.

488
00:28:55,875 --> 00:28:57,250
We are still on the way.

489
00:28:57,250 --> 00:28:59,541
It's a journey. We’re moving up.

490
00:28:59,541 --> 00:29:02,333
I think, you know,
I think genetics pay off in the 

491
00:29:02,333 --> 00:29:04,125
pharma space has been pretty clear.

492
00:29:04,125 --> 00:29:06,916
I think it's, Matthew Wilson.

493
00:29:06,916 --> 00:29:10,916
I shouldn't say the name,
but I think his publication outlined that.

494
00:29:11,375 --> 00:29:14,375
When you have genetic evidence
going into a program,

495
00:29:14,500 --> 00:29:20,333
you're more than twice as likely to have
a successful exit of that of that target.

496
00:29:20,750 --> 00:29:24,541
So I think we're still early days
with proteomics, but I'm very optimistic

497
00:29:24,541 --> 00:29:27,583
that having proteomics
evidence will further help us with

498
00:29:28,166 --> 00:29:32,166
with demonstrating that, 
a program is likely to be successful.

499
00:29:32,166 --> 00:29:33,166
So it's

500
00:29:33,166 --> 00:29:37,583
then we'll have to be able to juggle
all these hugely successful programs

501
00:29:37,583 --> 00:29:40,583
and get them out into the market
with the health care system

502
00:29:40,916 --> 00:29:43,250
that maybe unprepared to
pay for them. But we'll see.

503
00:29:44,208 --> 00:29:46,875
But that's, that's

504
00:29:46,875 --> 00:29:50,625
different problems for
different health care systems.

505
00:29:50,833 --> 00:29:51,250
But yeah.

506
00:29:51,250 --> 00:29:55,041
So so both of you, I'd love to understand
where you see

507
00:29:55,666 --> 00:29:59,750
an ability to have a subset of proteins
that really help us understand

508
00:29:59,750 --> 00:30:04,666
biological age and how biological age
may not be reflective

509
00:30:04,666 --> 00:30:08,791
of chronological age,
how might that actually be useful

510
00:30:08,791 --> 00:30:10,458
in the future as a clinical tool, 

511
00:30:10,458 --> 00:30:11,750
It's a great point.

512
00:30:11,750 --> 00:30:14,500
as a direct to consumer tool?

513
00:30:14,500 --> 00:30:17,375
If the ancestry.coms or 23andMe's of the world

514
00:30:17,375 --> 00:30:19,500
build something like this, how might people use it?

515
00:30:19,500 --> 00:30:21,333
What are your thoughts there?

516
00:30:21,333 --> 00:30:22,583
And also to add something here before 

517
00:30:22,583 --> 00:30:26,416
Cornelia, you're of course 
the best person to answer this,

518
00:30:26,416 --> 00:30:29,708
but also to add the fact that now we're not 
talking about single proteins or single genes,

519
00:30:29,708 --> 00:30:33,125
we're talking about pathways, 
we're talking about signatures at the end.

520
00:30:33,125 --> 00:30:36,416
And, we see a lot of inflammation
coming with aging.

521
00:30:36,958 --> 00:30:39,541
And I think probably

522
00:30:39,541 --> 00:30:42,833
we have to deep dive a little bit more 
in inflammation mechanism to understand aging.

523
00:30:42,833 --> 00:30:45,833
But yeah, I'm happy to hear your thoughts
how you see going to the clinics

524
00:30:45,833 --> 00:30:49,125
or how do you see go to the prognosis,
for example, from your prospective.

525
00:30:49,458 --> 00:30:52,250
Well, I think well, again,
we learn from the genetics.

526
00:30:52,250 --> 00:30:56,125
I think the 23andMe people are interested in

527
00:30:56,125 --> 00:31:00,333
in their genes,
either at the risk of the disease,

528
00:31:00,500 --> 00:31:02,166
but it was also in their heritage.

529
00:31:02,166 --> 00:31:07,708
I think if you look in, the UK,
we have the ZOE program

530
00:31:07,708 --> 00:31:11,625
where people I'm very much interested
in their microbiome.

531
00:31:12,125 --> 00:31:14,083
Again, it's a field in action.

532
00:31:14,083 --> 00:31:17,583
I can't believe that,
people getting the tools

533
00:31:17,583 --> 00:31:20,833
and the final tools in there,
but they get an impression

534
00:31:21,250 --> 00:31:24,333
how well their gut microbiome is functioning

535
00:31:24,833 --> 00:31:28,291
based on the state of the art
and and the truth on that.

536
00:31:28,583 --> 00:31:36,000
So I, I definitely think
that in the direct consumer field,

537
00:31:36,000 --> 00:31:37,125
this is exciting.

538
00:31:37,125 --> 00:31:40,125
This will be interesting.

539
00:31:40,125 --> 00:31:43,791
I can imagine that if you link
your microbiome to

540
00:31:44,125 --> 00:31:48,583
your aging profile that,
that it's even going be more interesting.

541
00:31:49,041 --> 00:31:52,041
And that is where I see
the field also going.

542
00:31:52,833 --> 00:31:58,458
What we trying to do
is starting out with the smoking data

543
00:31:59,041 --> 00:32:01,375
What we have to try out now is

544
00:32:01,375 --> 00:32:05,333
to what extent you can 
revert back your aging profile.

545
00:32:05,666 --> 00:32:08,666
And to me,

546
00:32:09,791 --> 00:32:13,583
based on what my gut feeling is

547
00:32:13,583 --> 00:32:16,083
 in there specifically,

548
00:32:16,083 --> 00:32:19,708
is that you probably can hold the processes

549
00:32:19,708 --> 00:32:22,458
as long as you intervene early.

550
00:32:22,458 --> 00:32:27,375
And old age, it's not clear,
but I think we have to find that out now.

551
00:32:27,708 --> 00:32:30,708
We don't know. Does it pay the price?

552
00:32:30,708 --> 00:32:33,958
If you are 90 plus to start doing
physical activity.

553
00:32:34,333 --> 00:32:39,166
Well, you ask me,
there's also dangers associated with it.

554
00:32:39,166 --> 00:32:42,166
I mean, we all know that
if your hip breaks,

555
00:32:42,166 --> 00:32:44,958
you have a broken hip after the age of 85,

556
00:32:44,958 --> 00:32:47,958
it's one of the strongest 
predictors of dying.

557
00:32:48,166 --> 00:32:52,916
But I think that
is what we are facing at the

558
00:32:53,916 --> 00:32:56,875
I think, well, the beauty is of
our analysis,

559
00:32:56,875 --> 00:33:00,250
it will give you a readout of interventions
that we always missed.

560
00:33:00,875 --> 00:33:03,583
I mean, if one of the interventions
that has been

561
00:33:03,583 --> 00:33:08,625
well pursuited is of course,
chlorectristration.

562
00:33:08,625 --> 00:33:13,083
Now, we all know that
that is quite a harsh job,

563
00:33:13,083 --> 00:33:16,083
because you really have to eat
less than you're supposed to eat

564
00:33:17,250 --> 00:33:19,791
Lika a third or something.

565
00:33:19,791 --> 00:33:21,250
It is quite harsh.

566
00:33:21,250 --> 00:33:26,083
And it really goes to this idea
that small animals live longer,

567
00:33:26,166 --> 00:33:29,166
than large animals.

568
00:33:30,958 --> 00:33:32,916
Really small men and women live longer,

569
00:33:32,916 --> 00:33:34,208
than tall men than women.

570
00:33:34,208 --> 00:33:39,875
And, there is a point to that and,
that is really targeted at this system.

571
00:33:39,875 --> 00:33:41,833
It's IGF one signaling.

572
00:33:41,833 --> 00:33:45,458
And in all animals
that is a problem

573
00:33:46,041 --> 00:33:47,666
for living long.

574
00:33:47,666 --> 00:33:50,833
So I think that is one of the outcomes.

575
00:33:50,833 --> 00:33:53,625
But I think it gives us hands and feet now,

576
00:33:53,625 --> 00:33:55,833
to have a readout that

577
00:33:56,916 --> 00:33:59,833
think about the monkey studies,

578
00:33:59,833 --> 00:34:01,875
in caloric restriction.

579
00:34:01,875 --> 00:34:03,583
There's only three, four done.

580
00:34:03,583 --> 00:34:07,541
You have to wait for ages 
before these monkeys age.

581
00:34:07,541 --> 00:34:10,875
And now we have a readout
that that is a little bit closer

582
00:34:11,333 --> 00:34:15,291
The readout seems 
to work already by age 40,

583
00:34:15,291 --> 00:34:18,291
and probably also age 20, 30.

584
00:34:18,708 --> 00:34:23,125
So hey, that must accelerate research also.

585
00:34:23,125 --> 00:34:26,125
And it must give us 
an insight whether intervention

586
00:34:26,458 --> 00:34:30,750
stopping smoking, 
don't wait for it just do it.

587
00:34:30,875 --> 00:34:35,000
Too much alcohol.
Stop that too.

588
00:34:35,000 --> 00:34:40,250
But physical activity was
if you talk to people in the aging field,

589
00:34:40,250 --> 00:34:43,583
some people are saying, well,
maybe good, but wait a minute,

590
00:34:43,583 --> 00:34:48,791
 if you're doing other physical activity,
also generating a lot of oxidative stress

591
00:34:48,791 --> 00:34:52,875
is that not also cause of aging?

592
00:34:52,875 --> 00:34:54,750
So I think we read it out now.

593
00:34:54,750 --> 00:34:56,166
We can read it out.

594
00:34:56,166 --> 00:34:58,041
It doesn't look that way in our hands.

595
00:34:58,041 --> 00:35:02,625
So it means that totally, you know,
some physical activity is good, and

596
00:35:02,666 --> 00:35:06,291
at least also for not only for vascular
but also for the brain.

597
00:35:06,291 --> 00:35:11,791
And I think that kind of opportunities,
the multitude to use it now as an outcome.

598
00:35:12,166 --> 00:35:17,208
We have to prove it 
but it looks that way that it is working.

599
00:35:18,291 --> 00:35:20,000
Well, you hear it here.

600
00:35:20,000 --> 00:35:22,125
Smoking, stop smoking,

601
00:35:22,125 --> 00:35:25,750
drink less alcohol, eat less food,

602
00:35:25,750 --> 00:35:30,416
and do exercise, 
but not to the extreme, right?

603
00:35:30,416 --> 00:35:32,916
Well,
but going back to the point of Sarantis,

604
00:35:32,916 --> 00:35:38,000
I think that inflammation 
we're all interested in it.

605
00:35:38,041 --> 00:35:40,291
But we also get now other proteins.

606
00:35:40,291 --> 00:35:43,375
That's also interesting and,

607
00:35:43,666 --> 00:35:46,625
what is the other thing
that is pushing us.

608
00:35:46,625 --> 00:35:50,750
And I definitely think that this was the start
for a lot of diseases

609
00:35:50,750 --> 00:35:54,291
and aging,
but also age-related diseases,

610
00:35:54,833 --> 00:35:58,333
but also exposures,
you know, the plastic exposure.

611
00:35:58,458 --> 00:35:59,916
Nobody knows what it does.

612
00:35:59,916 --> 00:36:01,458
I used to

613
00:36:01,458 --> 00:36:06,083
like if you have a readout for that,
that will inform us a little bit

614
00:36:06,083 --> 00:36:09,458
what goes on in the body
and how worried we should be.

615
00:36:09,458 --> 00:36:13,125
Yeah, PFOS, PFAS,
these sort of forever molecules.

616
00:36:13,125 --> 00:36:13,708
Would you.

617
00:36:13,708 --> 00:36:14,541
Would you like to comment,

618
00:36:14,541 --> 00:36:17,666
a little bit about the drug interventions
I mean old drugs.

619
00:36:17,666 --> 00:36:20,541
Old dog, new tricks, 
like rapamycin for example.

620
00:36:20,541 --> 00:36:23,416
Hg2 inhibitors, now we hear
that they are player or...

621
00:36:23,416 --> 00:36:24,791
What is your feeling about that?

622
00:36:24,791 --> 00:36:27,625
Targeting everything 
is targeting aging actually?

623
00:36:27,625 --> 00:36:28,875
Or vice versa?

624
00:36:28,875 --> 00:36:30,875
Why do you mention this, Sarantis?

625
00:36:30,875 --> 00:36:32,416
Because we won’t need to study that.

626
00:36:32,416 --> 00:36:37,583
So we have this week a break for 
what is our low hanging fruit.

627
00:36:37,583 --> 00:36:41,958
Because I knew if you join this field it's
not for the faint hearted.

628
00:36:41,958 --> 00:36:44,958
There's big competition, stiff competition
that we usually,

629
00:36:45,291 --> 00:36:47,666
we've always been reasonable about it,

630
00:36:47,666 --> 00:36:50,583
that we say, okay, if we see already a
publication.

631
00:36:50,583 --> 00:36:52,708
What is our lease?
What is are what is the low hanging fruit?

632
00:36:52,708 --> 00:36:57,083
And we definitely have
everything lined up there

633
00:36:57,083 --> 00:37:00,375
with Sihao
and Austin to do this aging clock.

634
00:37:00,791 --> 00:37:04,250
So but one of the things
that we are getting moving to

635
00:37:04,250 --> 00:37:08,750
as a field, of interest
is also the clinical application.

636
00:37:08,750 --> 00:37:15,541
We have already done a study 
that liver and alcohol are big problem.

637
00:37:15,541 --> 00:37:18,500
A big problem is also
that people don't know

638
00:37:18,500 --> 00:37:23,416
how much alcohol they use, and they don't
want to know how much alcohol they use.

639
00:37:23,416 --> 00:37:24,750
And the produce.

640
00:37:24,750 --> 00:37:30,083
So can we just, 
distinguished for liver diseases

641
00:37:30,125 --> 00:37:35,375
can we not use this profile for that,
then predict how long we will do this?

642
00:37:35,375 --> 00:37:38,375
And I of course,
it's used lots of alcohol

643
00:37:38,541 --> 00:37:42,375
and you get the usual diseases,
but you also get the cirrhosis

644
00:37:42,375 --> 00:37:45,166
and you get liver cancer.

645
00:37:45,166 --> 00:37:48,416
so here you go
so that is what we take as a benchmark.

646
00:37:48,916 --> 00:37:51,625
The other benchmark 
we definitely we're to use is 

647
00:37:51,625 --> 00:37:57,625
how to, serve the certian drugs,
how do they act what we know that.

648
00:37:57,666 --> 00:38:00,125
But also what is that unexpected actions.

649
00:38:00,125 --> 00:38:02,416
So this will be negative side effects.

650
00:38:02,416 --> 00:38:06,125
But we all know that some drugs,
think about statins you know,

651
00:38:06,125 --> 00:38:10,166
there was time this is, 
we are working here in the group

652
00:38:10,166 --> 00:38:11,541
that did the most statins research

653
00:38:11,541 --> 00:38:13,083
and you know except that

654
00:38:13,083 --> 00:38:17,083
some people get some muscle pain
and some very severe ones

655
00:38:17,083 --> 00:38:20,875
there is quite an argument to almost
put it in the drinking water, right.

656
00:38:20,875 --> 00:38:22,750
So of course you shouldn’t do that.

657
00:38:22,750 --> 00:38:27,041
But there's also positive effects, side
effects of the drugs which were never in

658
00:38:27,708 --> 00:38:32,208
the notes you get if you take the drugs
but it's very interesting.

659
00:38:32,208 --> 00:38:36,708
It's very interesting on this act 
for instance on inflammation and how

660
00:38:37,250 --> 00:38:41,083
so definitely
that is in part our target

661
00:38:41,083 --> 00:38:45,166
and that's also with the way
we're working population health and

662
00:38:46,583 --> 00:38:49,625
we should really resolve these issues.

663
00:38:50,375 --> 00:38:53,250
There's so much opportunity 
to understand mechanism,

664
00:38:53,250 --> 00:38:54,958
rapamycin,
like Serhant has mentioned,

665
00:38:54,958 --> 00:38:58,750
we don't really understand
metformin has some beneficial effects,

666
00:38:58,750 --> 00:39:04,083
but it can also alter
how exercise is, is benefiting us too.

667
00:39:04,083 --> 00:39:05,916
So understanding the mechanism of that,

668
00:39:05,916 --> 00:39:08,833
G... what are the GLP 1s?

669
00:39:09,208 --> 00:39:11,625
I mean, those are acting in the brain.

670
00:39:11,625 --> 00:39:12,791
That's fascinating. Right?

671
00:39:12,791 --> 00:39:16,916
We're really just parsing all that out
and it's already almost in the water

672
00:39:16,916 --> 00:39:17,875
for many.

673
00:39:17,875 --> 00:39:19,583
For many populations. Right.

674
00:39:19,583 --> 00:39:23,625
There's just so much opportunity
that I hope proteins can help.

675
00:39:23,625 --> 00:39:27,500
At least,
like I said, point to some hypotheses

676
00:39:27,500 --> 00:39:30,416
that can then be tested
by groups like yours, Cornelia

677
00:39:30,958 --> 00:39:34,000
So definitely that is a field of interest
and I.

678
00:39:34,000 --> 00:39:35,416
But on the other hand,

679
00:39:35,416 --> 00:39:39,041
the exposures of two exposures
to that shouldn’t be there.

680
00:39:39,166 --> 00:39:42,083
The plastics that are built,
the pesticides.

681
00:39:42,083 --> 00:39:43,083
I think we see them.

682
00:39:43,083 --> 00:39:46,083
We see that the, you know, that.

683
00:39:46,458 --> 00:39:50,083
And, the fact that there's 
air pollution in the region,

684
00:39:50,250 --> 00:39:53,666
pops up
as having the similar effect of smoking.

685
00:39:53,666 --> 00:39:55,458
And that is not good.

686
00:39:55,458 --> 00:39:59,500
So I think there's a lot of opportunities
and we need a lot of hands,

687
00:39:59,500 --> 00:40:02,416
but also a lot of brains to do that.

688
00:40:02,750 --> 00:40:04,000
And technologies.

689
00:40:04,000 --> 00:40:05,291
And technologies to do that.

690
00:40:05,291 --> 00:40:09,500
and that definitely,
we need more of the protein.

691
00:40:09,500 --> 00:40:11,208
We know that there’s lots more proteins.

692
00:40:11,208 --> 00:40:14,916
We need more, the different isoforms.

693
00:40:14,916 --> 00:40:18,875
We need to know more
about the phosphorylation and the,

694
00:40:19,333 --> 00:40:22,083
processes of processing of these proteins.

695
00:40:22,083 --> 00:40:25,250
But it it isn't that a field
that, you know, we

696
00:40:25,958 --> 00:40:30,750
I'm not I'm not young anymore,
but I think yeah, I think we definitely

697
00:40:31,208 --> 00:40:35,541
the future will tell a lot about what
we always have been wondering about.

698
00:40:37,250 --> 00:40:37,791
To that.

699
00:40:37,791 --> 00:40:40,625
To the point around
the needs for this area.

700
00:40:40,625 --> 00:40:43,250
What are the cohorts that come to mind

701
00:40:43,250 --> 00:40:46,125
that are collecting
environmental information

702
00:40:46,125 --> 00:40:49,875
that you think are ones
we want to highlight and promote?

703
00:40:50,208 --> 00:40:53,833
Because it's like I said, it's
hard to collect these sort

704
00:40:53,833 --> 00:40:55,250
of environmental variables.

705
00:40:55,250 --> 00:40:59,000
Are there ones that you particularly
like that you want to make sure,

706
00:40:59,541 --> 00:41:04,125
are successful in the future, continue
to collect data, that sort of thing?

707
00:41:04,125 --> 00:41:06,791
I think that there are many cohorts now.

708
00:41:06,791 --> 00:41:11,000
That,
of course, has, has really dedicated

709
00:41:11,000 --> 00:41:14,000
their life to look at multiple exposures.

710
00:41:14,166 --> 00:41:17,041
I really favor
the epidemiological setting.

711
00:41:17,041 --> 00:41:20,208
And the reason for that is that,

712
00:41:20,208 --> 00:41:23,875
what you probably, if you single out
one exposure,

713
00:41:23,875 --> 00:41:27,625
right, it's unlikely that in your life

714
00:41:27,625 --> 00:41:31,583
you only have one exposure
you need a broader picture.

715
00:41:31,958 --> 00:41:36,000
So I, I'm brought up in
a department

716
00:41:36,000 --> 00:41:41,041
in Rotterdam
where we always, looked and try

717
00:41:41,083 --> 00:41:46,291
to look at the complete picture
with the view that in the end of the day,

718
00:41:46,291 --> 00:41:49,000
you're asking yourself
what is the effect of smoking?

719
00:41:49,000 --> 00:41:52,041
Oh, but,
you know, if you smoke, you often

720
00:41:52,041 --> 00:41:55,875
more likely to drink a lot of coffee,
a lot of alcohol.

721
00:41:55,875 --> 00:41:58,875
You're more likely to use oral contraceptives.

722
00:41:59,208 --> 00:42:02,416
Hey, there’s a lot more things you do.

723
00:42:02,416 --> 00:42:07,500
And, I think that these studies
have been incredibly powerful.

724
00:42:07,500 --> 00:42:13,000
And incredibly important, the UK Biobank
is a is a fantastic example on that,

725
00:42:13,416 --> 00:42:18,083
that also data have been gathered,
you know, they been adding of data

726
00:42:18,083 --> 00:42:20,708
stacked onto each other.

727
00:42:20,708 --> 00:42:23,916
And that allows you to do
multi-omics studies

728
00:42:23,916 --> 00:42:28,083
in a very valid way,
but also weigh in exposures.

729
00:42:28,666 --> 00:42:32,625
Now, one of the examples
I would give that convinced me totally

730
00:42:32,625 --> 00:42:35,375
is that you have to look at, 
multi-omics.

731
00:42:35,375 --> 00:42:40,583
Is that what, we did
is look at metabolomics,

732
00:42:40,583 --> 00:42:42,500
and we started thinking, why?

733
00:42:42,916 --> 00:42:47,750
Well, the idea is metabolomics
is genetically determined, but so environment,

734
00:42:47,750 --> 00:42:51,250
is the active component?

735
00:42:51,541 --> 00:42:57,291
And you're really getting quite overwhelmed

736
00:42:57,291 --> 00:43:01,958
if you look at the how strong medication
also influences metabolomics.

737
00:43:02,458 --> 00:43:06,833
We're now going back the same 
as Sarantis on proteomics

738
00:43:06,833 --> 00:43:15,958
and for some it’s really overwhelming
how it's medication is, influencing your proteome.

739
00:43:16,125 --> 00:43:18,666
Now look in the most of the epidemiologists

740
00:43:18,666 --> 00:43:21,666
have been wise
and have been gathering data

741
00:43:21,666 --> 00:43:25,500
of a lot of exposures
and that will be helpful.

742
00:43:26,083 --> 00:43:31,291
And definitely the medication
you need to take that into account.

743
00:43:31,291 --> 00:43:33,833
But, on the other hand,
they should look at medication.

744
00:43:33,833 --> 00:43:38,583
The smokers also turned off to be, 
a confounding factor for that.

745
00:43:39,250 --> 00:43:42,166
But, you know,
the fact that both metabolomics

746
00:43:42,166 --> 00:43:46,291
but also proteomics
even more is associated to medication,

747
00:43:46,333 --> 00:43:49,333
suggests what we already have hypothesized

748
00:43:49,375 --> 00:43:53,791
that a lot of medication
is somehow targeting proteome.

749
00:43:54,416 --> 00:43:55,000
Yeah. It's.

750
00:43:55,000 --> 00:43:58,250
It's the messy part of the data. Right?

751
00:43:58,250 --> 00:44:01,250
But it's because we are collecting it across

752
00:44:02,583 --> 00:44:06,083
ideally large numbers of people
that signal can emerge

753
00:44:06,375 --> 00:44:10,125
even even though there's challenges
in collecting those data.

754
00:44:10,458 --> 00:44:13,291
I think more and more we should include

755
00:44:13,291 --> 00:44:16,333
it also proteomics in in trials.

756
00:44:16,500 --> 00:44:17,750
We should do that.

757
00:44:17,750 --> 00:44:21,250
And it's in clinical trials
in which you test medication.

758
00:44:21,833 --> 00:44:25,791
But please if we do
these intervention trials also show me

759
00:44:25,791 --> 00:44:29,625
that you have an effect of the proteins
that develop the disease.

760
00:44:30,125 --> 00:44:34,041
and there is our aging work is important,
but there's a lot of more,

761
00:44:34,125 --> 00:44:38,625
profiles that we need for dementia
in the early phase.

762
00:44:38,625 --> 00:44:43,166
So not the fact that you have P tau,
which is just a signal that your head is

763
00:44:43,166 --> 00:44:45,375
full of tau and if your head if full of tau...

764
00:44:45,583 --> 00:44:48,583
It’s one only biomarker, right?

765
00:44:48,708 --> 00:44:49,750
We need something earlier.

766
00:44:50,583 --> 00:44:51,708
We need more.

767
00:44:51,708 --> 00:44:55,041
I don't think that if, physical activity

768
00:44:55,041 --> 00:44:59,625
protects you against dementia,
you shouldn’t start with it at age 85.

769
00:44:59,625 --> 00:45:01,666
You should start with that early.

770
00:45:01,666 --> 00:45:04,333
And we've I've read out of studies
that show death.

771
00:45:04,333 --> 00:45:06,791
That convinced me. And of course, the...

772
00:45:06,791 --> 00:45:09,916
Yeah, a little advice
is that we have on the exposures

773
00:45:10,958 --> 00:45:13,166
are interesting, but we need much more.

774
00:45:13,166 --> 00:45:14,500
We need much more.

775
00:45:14,500 --> 00:45:18,708
On nowadays what
we are exposed to that even the fact that

776
00:45:18,708 --> 00:45:23,375
our sleep is different, that
we are exposed to light at night,

777
00:45:23,375 --> 00:45:25,291
that we never were exposed to.

778
00:45:26,166 --> 00:45:27,708
There's a lot to be learned.

779
00:45:27,708 --> 00:45:30,708
And I think that type of trials,

780
00:45:30,708 --> 00:45:34,875
there’s two things on trials for exposures,
is the first of all,

781
00:45:34,875 --> 00:45:38,500
they have to be big,
even for caloric restrictions.

782
00:45:38,500 --> 00:45:41,500
You see all these smaller studies,
people lose weight.

783
00:45:41,500 --> 00:45:43,791
I mean,
we have the better outcome, right?

784
00:45:43,791 --> 00:45:46,583
of course we lose weight
if you don't eat the calories.

785
00:45:46,583 --> 00:45:49,875
it is obvious
that that will happen.

786
00:45:49,875 --> 00:45:53,541
But we need the readouts of that,
that shows us

787
00:45:53,833 --> 00:45:58,666
that it takes
really that it stops aging.

788
00:45:58,666 --> 00:46:03,375
And the trials, I came to Oxford

789
00:46:03,375 --> 00:46:07,083
to the Oxford department, partly

790
00:46:07,125 --> 00:46:10,250
because the trials are so big,
but partly I like the spirit

791
00:46:10,250 --> 00:46:14,625
about the trials here, that they have
to be big in order to show things,

792
00:46:15,083 --> 00:46:18,083
because that affects sometimes I, I mean,

793
00:46:18,833 --> 00:46:22,750
are still subtle. I think we got used
to that in the genetics too.

794
00:46:23,375 --> 00:46:25,833
Of course you have genes with big effects,

795
00:46:25,833 --> 00:46:28,250
but a lot of them
will not have that big effect.

796
00:46:28,250 --> 00:46:32,750
It's the aggregate of all the genes
and if it's the aggregate of the genes,

797
00:46:32,750 --> 00:46:35,125
it has to be
the aggregate of the proteins to.

798
00:46:36,250 --> 00:46:39,583
Otherwise the effect of these
all these genes don't make sense.

799
00:46:39,583 --> 00:46:43,625
So I think that is what we’re facing that,

800
00:46:43,666 --> 00:46:46,791
we have to start thinking of trials
with complex outcomes.

801
00:46:46,791 --> 00:46:49,791
And we have had a lot of benefit

802
00:46:50,083 --> 00:46:57,291
that coming to Oxford I really wanted
to start looking at machine learning too.

803
00:46:57,291 --> 00:46:59,916
And that gave us also a

804
00:47:00,833 --> 00:47:03,000
very much of a boost, I should say.

805
00:47:03,000 --> 00:47:07,875
I'm not saying that machine learning
solves everything, and a definitely not.

806
00:47:07,875 --> 00:47:09,541
You don't hear me say that.

807
00:47:09,541 --> 00:47:13,791
But if you look at the in a simple,
even simple

808
00:47:13,791 --> 00:47:17,875
machine learning models, it can deal
with the complexity a bit easier.

809
00:47:17,875 --> 00:47:20,875
And I think we we nailed that down.

810
00:47:22,125 --> 00:47:25,583
And for strong associations
like the proteomics age group,

811
00:47:25,875 --> 00:47:30,125
it really doesn't matter what you take
a more classical approach

812
00:47:30,125 --> 00:47:34,541
like elastic net or gradient boosting,
which is kind of a random forest

813
00:47:34,541 --> 00:47:38,833
or you take a neural network,
but in the end of the day,

814
00:47:38,833 --> 00:47:44,000
it may be that that some of these methods
may be more powerful

815
00:47:44,208 --> 00:47:49,416
to pick up these aggregates and also
translate it back that you get into your hands.

816
00:47:50,500 --> 00:47:52,666
Which plotting is doing what?

817
00:47:52,666 --> 00:47:55,916
If it becomes completely obscure
in the neural network,

818
00:47:55,916 --> 00:47:59,458
what has done what?
Are you really going to invest

819
00:47:59,458 --> 00:48:03,375
hundreds of millions
to develop in therapy for that?

820
00:48:03,375 --> 00:48:05,875
No, you want to first know,

821
00:48:05,875 --> 00:48:07,333
not too many proteins,

822
00:48:07,333 --> 00:48:10,000
and what is doing what, tell me. Right?

823
00:48:10,208 --> 00:48:12,083
And that is when you have to be able to start out.

824
00:48:12,083 --> 00:48:15,166
And machine learning, it's giving us a lot.

825
00:48:16,041 --> 00:48:17,375
Yeah, yeah. And.

826
00:48:17,375 --> 00:48:19,500
But we have to be careful
about overtraining.

827
00:48:19,500 --> 00:48:22,000
But that's where having this

828
00:48:22,000 --> 00:48:25,791
growing field of 
machine learning is informing us.

829
00:48:25,791 --> 00:48:29,833
But I think parsing out, 
what's the genetic contribution from ancestry?

830
00:48:29,833 --> 00:48:31,583
What's the contribution from gender?

831
00:48:31,583 --> 00:48:31,750
Yeah.

832
00:48:31,916 --> 00:48:35,500
What are the signals 
in the proteins that confer gender

833
00:48:35,500 --> 00:48:37,291
that you can then use to stratify that?

834
00:48:37,291 --> 00:48:41,958
There's so much complexity 
that machine learning is helping us to parse out.

835
00:48:41,958 --> 00:48:42,958
Yeah, and that we were lucky.

836
00:48:42,958 --> 00:48:46,750
I think we've been wonderful here in Oxford.

837
00:48:46,750 --> 00:48:48,875
That we have multiple
all these cohorts

838
00:48:48,875 --> 00:48:51,333
We have the China Kadoorie Biobank, 

839
00:48:51,750 --> 00:48:52,166
Yeah.

840
00:48:52,166 --> 00:48:55,166
you saw that it’s fantastic
what they are setting up.

841
00:48:55,833 --> 00:49:00,291
We have bigger cohorts in the million
women study, but we have also,

842
00:49:00,666 --> 00:49:04,041
the large trials that have been done.

843
00:49:04,041 --> 00:49:08,125
And, you know, even in a trial,
you can do now, start thinking

844
00:49:08,125 --> 00:49:12,083
of a silico experiments
that if the trial has been done

845
00:49:12,083 --> 00:49:16,291
with a certain drug
that you want to repurpose,

846
00:49:16,291 --> 00:49:20,958
you can just measure in that trial
what the effects of the proteins are.

847
00:49:21,416 --> 00:49:23,833
I think you really have to go.

848
00:49:23,833 --> 00:49:25,583
We have to be intelligible.

849
00:49:25,583 --> 00:49:28,458
And more intelligent
on how to repurpose,

850
00:49:28,458 --> 00:49:31,375
and reuse the studies that we had.

851
00:49:31,375 --> 00:49:34,791
But the fact I,
I totally agree with people that say,

852
00:49:34,791 --> 00:49:37,791
if you split two data in the training
and the test set,

853
00:49:38,291 --> 00:49:40,916
if there's structure in your data,

854
00:49:40,916 --> 00:49:43,916
then it’s in your training
and your test set and then

855
00:49:43,916 --> 00:49:47,416
in my early days using machine learning

856
00:49:47,416 --> 00:49:53,083
in team discovery is that,
we figured out the hard way.

857
00:49:53,083 --> 00:49:57,541
We had the test set
and training set replicated,

858
00:49:57,916 --> 00:50:00,708
but when we finally looked, 

859
00:50:00,708 --> 00:50:03,708
what the neural network was using,

860
00:50:03,833 --> 00:50:08,208
it was using missing data to predict
things like, how is it possible?

861
00:50:08,208 --> 00:50:11,333
How is that possible
that you can predict with missing data?

862
00:50:11,708 --> 00:50:13,833
There must be something that you can’t input

863
00:50:13,833 --> 00:50:17,166
in that range region well,
or there must be a reason for that.

864
00:50:17,166 --> 00:50:21,291
But if the problem with the missing
data is in your training set

865
00:50:21,291 --> 00:50:22,791
it’s also in your test set.

866
00:50:22,791 --> 00:50:26,000
So for us, it's
so important that you can use

867
00:50:26,000 --> 00:50:30,958
data across studies that we could use UK
Biobank as a powerhouse

868
00:50:31,583 --> 00:50:32,750
and a powerful tool,

869
00:50:32,750 --> 00:50:37,333
but that we can replicate it in other studies

870
00:50:37,333 --> 00:50:41,333
that are completely independent
and that will be important in genetics.

871
00:50:41,333 --> 00:50:44,333
It will also be important in exposures.

872
00:50:44,750 --> 00:50:49,000
And there are more and more studies
that are integrating, I know of,

873
00:50:49,000 --> 00:50:52,041
Olink proteomics,
of course, that are integrating these data

874
00:50:52,041 --> 00:50:55,750
with genetic data that will offer
opportunities for collaboration.

875
00:50:55,750 --> 00:50:57,625
Yeah. Awesome.

876
00:50:57,625 --> 00:51:02,083
Well, so I think this is a great place
for us to sort of wrap up.

877
00:51:02,083 --> 00:51:06,583
I'd love to give you a chance to say any
last thoughts that you'd like to share?

878
00:51:06,833 --> 00:51:07,583
Cornelia.

879
00:51:08,625 --> 00:51:11,625
No. Anything that it’s...

880
00:51:11,750 --> 00:51:13,250
The only take home is

881
00:51:13,250 --> 00:51:16,333
well, I already mentioned
the race is not the realm.

882
00:51:16,333 --> 00:51:18,916
I think it's not technology.

883
00:51:18,916 --> 00:51:23,416
You guys are still finding,
better ways to quantify the protein.

884
00:51:23,708 --> 00:51:28,583
You are finding better ways 
to describe the proteome.

885
00:51:29,375 --> 00:51:31,333
That will be ongoing.

886
00:51:31,333 --> 00:51:34,333
I think it's quite exciting
to be in this field.

887
00:51:34,666 --> 00:51:39,708
The other way around is that for us, 
in the data science and

888
00:51:39,708 --> 00:51:42,458
the epidemiology,
I think there's a lot of work to do.

889
00:51:42,458 --> 00:51:45,041
A lot of thinking to do 
how to analyze the data,

890
00:51:45,041 --> 00:51:49,041
how to integrate it 
over the exposome, the genome.

891
00:51:49,041 --> 00:51:53,083
And then, it's
going to be very exciting on that field

892
00:51:53,166 --> 00:51:58,583
telling people how to prevent the disease
better, giving them tools to monitor.

893
00:51:58,583 --> 00:52:02,875
And nobody would have thought,
 20 years ago

894
00:52:02,916 --> 00:52:05,500
first of all, 
that we were going not outside

895
00:52:05,500 --> 00:52:08,000
of or house
without taking the phone.

896
00:52:08,416 --> 00:52:12,166
But, you know,
that we would be having Apple Watches.

897
00:52:12,166 --> 00:52:13,625
and Fitbits

898
00:52:14,291 --> 00:52:16,583
Yeah, how many computers do we carry?

899
00:52:16,583 --> 00:52:20,000
For them it was difficult to understand
that we landed on the moon.

900
00:52:20,000 --> 00:52:22,541
And to accept that.

901
00:52:22,541 --> 00:52:27,458
But you know nowadays
this is the field that’s going to develop

902
00:52:27,458 --> 00:52:31,833
and we’re going to be boosted also 
on the data analytics,

903
00:52:31,833 --> 00:52:34,208
on the integration of data,

904
00:52:34,541 --> 00:52:39,833
the use of machine learning,
the abuse, but also correct that again

905
00:52:39,833 --> 00:52:41,875
And how to translate that back.

906
00:52:42,166 --> 00:52:45,000
You know, it's not only the data science
that is relevant.

907
00:52:45,000 --> 00:52:47,250
In the end of the day, it's relevant,

908
00:52:47,250 --> 00:52:51,916
what you do, the impact that you have in
in curing people,

909
00:52:52,291 --> 00:52:53,916
in presenting the disease

910
00:52:53,916 --> 00:52:57,833
because you know, if anything in your life
you don't will to become diseased,

911
00:52:57,833 --> 00:53:01,000
you want to prevent it. 
Definitely with dementia,

912
00:53:01,000 --> 00:53:03,083
but also with many other diseases.

913
00:53:04,208 --> 00:53:06,625
It's that translational that counts.

914
00:53:06,625 --> 00:53:08,416
And that is important.

915
00:53:08,416 --> 00:53:11,958
And it's important
that we all keep that in mind.

916
00:53:13,250 --> 00:53:14,208
Couldn't agree more.

917
00:53:14,208 --> 00:53:15,541
And you all heard it here,

918
00:53:15,541 --> 00:53:16,666
this is the place to go,

919
00:53:16,666 --> 00:53:19,125
Oxford is the place to go for large
data sets.

920
00:53:19,125 --> 00:53:24,375
They're amazing cohorts here 
and amazing scientists to work with.

921
00:53:24,416 --> 00:53:28,666
So for those of you who are thinking
about post-docs or PhDs, think about that.

922
00:53:28,875 --> 00:53:31,541
Sarantis, I'll give you a chance to please

923
00:53:31,541 --> 00:53:32,083
It was great.

924
00:53:32,083 --> 00:53:33,166
Any last thought?

925
00:53:33,166 --> 00:53:36,625
It was great to hear from Cornelia, 
about the aging and aging-related diseases

926
00:53:36,625 --> 00:53:39,291
I think proteins play 
a really important role on that.

927
00:53:39,291 --> 00:53:43,791
Plasma proteome is on spot now 
and we can,

928
00:53:43,791 --> 00:53:47,875
using this plasma proteome,
we can understand the biology of disease

929
00:53:47,875 --> 00:53:49,666
from different tissue types.

930
00:53:49,666 --> 00:53:53,416
We've also we can also understand
of different tissue types, phenotypes

931
00:53:53,500 --> 00:53:55,250
screening only for plasma proteomics.

932
00:53:55,250 --> 00:53:56,958
I think that's the take home message here.

933
00:53:56,958 --> 00:53:59,458
And really nice
to have you, Cornelia.

934
00:53:59,458 --> 00:54:00,791
Great, I enjoyed it a lot.

935
00:54:00,791 --> 00:54:02,041
Thank you very much, Cindy.

936
00:54:02,041 --> 00:54:04,750
And yeah I mean the
the last word is for you, Cindy.

937
00:54:05,250 --> 00:54:06,666
Super fun. Super fun!

938
00:54:06,666 --> 00:54:09,541
So then
I'll just go back and double click on.

939
00:54:09,541 --> 00:54:14,458
So I mentioned, the study that
that demonstrated that having genetic data

940
00:54:14,458 --> 00:54:18,583
going into a clinical trial helps
improve success by at least two times.

941
00:54:18,958 --> 00:54:20,708
That was actually Matt Nelson.

942
00:54:20,708 --> 00:54:22,500
So apologies for that, 2015,

943
00:54:22,500 --> 00:54:25,291
and that was a Nature Genetics paper
really pivotal paper.

944
00:54:25,291 --> 00:54:28,250
And then of course
AstraZeneca has also published

945
00:54:28,250 --> 00:54:32,333
on their ways of filtering and leveraging
genetic data in different ways.

946
00:54:32,333 --> 00:54:34,041
Gives them a seven times

947
00:54:34,041 --> 00:54:38,333
improvement in clinical trial outcomes,
which is which I just wanted to highlight.

948
00:54:38,625 --> 00:54:41,333
And then we also mentioned Austin Argentieri.

949
00:54:41,333 --> 00:54:46,375
We mentioned, Sihao Zhao, a PhD student
and then we also talked

950
00:54:46,375 --> 00:54:49,708
about China Kadoorie Biobank,
but we didn't mention Zhengming.

951
00:54:49,708 --> 00:54:55,958
So I want to I want to, give a shout out
to the amazing biobank that he's built.

952
00:54:55,958 --> 00:55:00,375
As I understand it, really,
a lot of the UK Biobank structure,

953
00:55:00,416 --> 00:55:05,125
was founded in how Zhengming, 
built out the China Kadoorie Biobank.

954
00:55:05,125 --> 00:55:06,833
So those two are great ones.

955
00:55:06,833 --> 00:55:10,541
And people use them a lot
for corroboration and combining data.

956
00:55:11,000 --> 00:55:15,375
And Austin is a great example
of someone who's done that so well.

957
00:55:15,375 --> 00:55:17,416
Well, in the future. Get him.

958
00:55:17,416 --> 00:55:18,750
Get him on the podcast.

959
00:55:18,750 --> 00:55:22,208
Perhaps once that paper comes out
and that paper will probably be out

960
00:55:22,208 --> 00:55:26,583
by the time we get this podcast,
published, I hope so.

961
00:55:26,583 --> 00:55:27,500
We can use this.

962
00:55:27,500 --> 00:55:31,416
This is an opportunity
to promote that important work.

963
00:55:31,416 --> 00:55:36,916
And so with with all of that exciting, 
content that we've talked about today.

964
00:55:36,916 --> 00:55:40,833
And I want to thank you, Cornelia,
so much for agreeing to

965
00:55:40,875 --> 00:55:43,208
to come on and trust us with
some of your story.

966
00:55:44,541 --> 00:55:45,583
Thank you very much.

967
00:55:46,166 --> 00:55:47,750
Thank you for having me.

968
00:55:51,125 --> 00:55:54,666
Well, that wraps up
this episode of Proteomics in Proximity.

969
00:55:55,166 --> 00:55:59,375
Huge thanks to our guests and authors
of such impactful publications.

970
00:55:59,708 --> 00:56:02,250
I also want to thank you for tuning in.

971
00:56:02,250 --> 00:56:04,541
Really appreciate you being here.

972
00:56:04,541 --> 00:56:06,333
If you enjoyed the content of this

973
00:56:06,333 --> 00:56:10,208
episode, please think about sharing it
with friends or colleagues

974
00:56:10,208 --> 00:56:12,458
you think might be interested
in the content.

975
00:56:12,458 --> 00:56:16,541
In addition, if you'd be willing
to head over to Apple or Spotify

976
00:56:16,541 --> 00:56:20,000
or wherever you digest your podcasts
and give us a rating and review,

977
00:56:20,000 --> 00:56:22,041
this will help others find the podcast

978
00:56:22,041 --> 00:56:25,375
when they're searching for proteomics
or precision medicine podcasts.

979
00:56:25,583 --> 00:56:29,250
And mostly I want to say
we would love to hear from you.

980
00:56:29,333 --> 00:56:32,875
So we have a dedicated email address
pip@olink.com

981
00:56:32,875 --> 00:56:34,333
Please reach out.

982
00:56:34,333 --> 00:56:38,291
Let us know what you're interested
in hearing about, what you care about,

983
00:56:38,291 --> 00:56:42,333
and any feedback on the episodes
that we have already done so far.

984
00:56:42,625 --> 00:56:46,000
This is all about you,
and so we're really keen

985
00:56:46,000 --> 00:56:49,000
to make sure that we're meeting
what you like to hear about.

986
00:56:49,291 --> 00:56:51,375
Thank you so much and we'll see you soon.