1 00:00:04,750 --> 00:00:06,606 Welcome to the 2 00:00:06,628 --> 00:00:08,926 Proteomics in Proximity podcast, where 3 00:00:08,948 --> 00:00:10,986 your co-hosts, Dale Yuzuki, Cindy 4 00:00:11,018 --> 00:00:13,338 Lawley and Sarantis Chlamydus from Olink 5 00:00:13,354 --> 00:00:15,726 Proteomics talk about the intersection of 6 00:00:15,748 --> 00:00:17,962 Proteomics with genomics for drug target 7 00:00:18,026 --> 00:00:20,526 discovery, the application of proteomics to 8 00:00:20,548 --> 00:00:22,830 reveal disease biomarkers, and current 9 00:00:22,900 --> 00:00:24,960 trends in using proteomics to 10 00:00:25,044 --> 00:00:27,814 unlock biological mechanisms. Here we have 11 00:00:27,852 --> 00:00:29,942 your hosts, Dale, Cindy, and 12 00:00:29,996 --> 00:00:31,610 Sarantis. 13 00:00:31,610 --> 00:00:33,450 Hey, everyone. 14 00:00:33,450 --> 00:00:35,366 Welcome to Proteomics in 15 00:00:35,388 --> 00:00:37,778 Proximity. Today we have a guest, 16 00:00:37,874 --> 00:00:39,846 Chris Whelan, who's joining us from 17 00:00:39,868 --> 00:00:41,574 Janssen Pharmaceuticals. 18 00:00:41,702 --> 00:00:43,722 Chris is the one who has 19 00:00:43,776 --> 00:00:45,814 helped spearhead bringing proteomics 20 00:00:45,862 --> 00:00:47,754 into the UK Biobank. So we're super 21 00:00:47,792 --> 00:00:49,450 excited to talk to him about his history, 22 00:00:49,520 --> 00:00:51,718 his background, and what 23 00:00:51,904 --> 00:00:53,434 the vision of bringing 24 00:00:53,482 --> 00:00:55,534 proteomics together with the 25 00:00:55,572 --> 00:00:57,994 genetics that UK Biobank is so famous 26 00:00:58,042 --> 00:01:00,766 for, the genetics and clinical data that 27 00:01:00,788 --> 00:01:02,826 we're all very excited about on the UK 28 00:01:02,858 --> 00:01:04,366 Biobank Research Analysis 29 00:01:04,398 --> 00:01:06,514 platform. And this week is a pretty 30 00:01:06,552 --> 00:01:08,866 auspicious week because we've just heard 31 00:01:08,968 --> 00:01:10,926 that the first tranche of data from the UK 32 00:01:10,958 --> 00:01:12,494 Biobank Pharma Proteomics 33 00:01:12,542 --> 00:01:14,882 Project have become available 34 00:01:15,016 --> 00:01:17,106 through the Research Analysis platform. So 35 00:01:17,128 --> 00:01:19,366 we're excited to talk to Chris about that as 36 00:01:19,388 --> 00:01:21,862 well. So, welcome, Chris. Hey, 37 00:01:21,916 --> 00:01:23,686 Cindy, Dale, Sarantis. How are you all 38 00:01:23,708 --> 00:01:25,926 doing? Doing great. It's great to 39 00:01:25,948 --> 00:01:27,990 have you with us today. 40 00:01:27,990 --> 00:01:29,350 Welcome, 41 00:01:29,420 --> 00:01:31,790 Chris. 42 00:01:31,790 --> 00:01:33,962 So Chris, can you tell us a little bit about 43 00:01:34,016 --> 00:01:36,554 your background in terms of going into 44 00:01:36,592 --> 00:01:38,266 science? You don't have to start sort of in 45 00:01:38,288 --> 00:01:40,554 your elementary school days, but certainly 46 00:01:40,672 --> 00:01:42,954 sort of your path 47 00:01:43,082 --> 00:01:45,246 to industry because I think that's always an 48 00:01:45,268 --> 00:01:47,890 interesting place to start. 49 00:01:47,890 --> 00:01:49,900 Absolutely, yeah. Happy to. 50 00:01:49,900 --> 00:01:51,694 So I did all of my training 51 00:01:51,812 --> 00:01:53,854 up to getting my PhD in 52 00:01:53,892 --> 00:01:55,440 Dublin, Ireland. 53 00:01:55,650 --> 00:01:57,426 I've been told recently that I'm losing my 54 00:01:57,448 --> 00:01:58,786 accent, so I'm going to try to make an 55 00:01:58,808 --> 00:02:00,820 effort to sound more today. 56 00:02:00,820 --> 00:02:02,760 But yeah, I 57 00:02:02,888 --> 00:02:04,546 started off in psychology for my 58 00:02:04,568 --> 00:02:06,566 undergrad and then realized I wanted to get 59 00:02:06,588 --> 00:02:08,818 more into the cellular 60 00:02:08,914 --> 00:02:10,822 sort of, sort of hard 61 00:02:10,876 --> 00:02:12,866 science behind brain 62 00:02:12,898 --> 00:02:14,980 illnesses. So did my Masters 63 00:02:14,980 --> 00:02:16,870 and my Ph.D. in neuroscience. 64 00:02:16,870 --> 00:02:18,518 One of my advisors was a 65 00:02:18,524 --> 00:02:20,770 geneticist. So I started to dip my 66 00:02:20,770 --> 00:02:22,694 toes in statistical 67 00:02:22,742 --> 00:02:24,938 genetics. And that sort of 68 00:02:24,944 --> 00:02:26,778 led me towards my postdoc in 69 00:02:26,864 --> 00:02:28,438 Los Angeles with the ENIGMA 70 00:02:28,454 --> 00:02:30,894 Consortium at USC. So there they were 71 00:02:30,932 --> 00:02:32,494 combining neuroimaging with 72 00:02:32,532 --> 00:02:34,670 GWAS, basically running, 73 00:02:34,740 --> 00:02:36,714 genome-wide association 74 00:02:36,762 --> 00:02:38,766 studies on very large collections of 75 00:02:38,788 --> 00:02:40,926 MRI scans. So I did that for 76 00:02:40,948 --> 00:02:42,490 two years. I 77 00:02:42,610 --> 00:02:44,558 felt that I always thought that I would be 78 00:02:44,564 --> 00:02:46,846 on the academic track. I remember 79 00:02:46,948 --> 00:02:48,818 in my PhD class, they actually 80 00:02:48,904 --> 00:02:50,898 wanted to do a straw poll of who wants to 81 00:02:50,904 --> 00:02:52,866 go to industry and who wants to be a 82 00:02:52,888 --> 00:02:54,926 lecturer or a professor. And I was firmly 83 00:02:54,958 --> 00:02:56,550 in the professor camp. But 84 00:02:56,550 --> 00:02:58,840 I think 85 00:02:58,840 --> 00:03:00,966 two years in 86 00:03:00,988 --> 00:03:02,994 academia in the States, it's tough. 87 00:03:03,042 --> 00:03:05,254 It's tough. And I actually had a good 88 00:03:05,292 --> 00:03:07,894 postdoc. My P.I. was awesome. Really 89 00:03:07,932 --> 00:03:09,674 lovely, man. Really supportive. But it just 90 00:03:09,712 --> 00:03:11,878 gave me some insights into it. It's 91 00:03:11,894 --> 00:03:13,890 a tough place to be. 92 00:03:13,890 --> 00:03:15,910 Beyond that, I think 93 00:03:15,910 --> 00:03:17,994 I wanted 94 00:03:18,032 --> 00:03:20,426 to be closer to the patients. That might 95 00:03:20,448 --> 00:03:22,106 sound like a little bit of a cliche, but I 96 00:03:22,128 --> 00:03:24,266 wanted to be really working on whatever I'm 97 00:03:24,298 --> 00:03:26,686 doing. I can see this affecting a 98 00:03:26,708 --> 00:03:28,862 patient in six or seven years 99 00:03:28,916 --> 00:03:30,958 time. So, I was going to 100 00:03:30,964 --> 00:03:32,862 move home to Dublin and then I 101 00:03:32,916 --> 00:03:34,750 got the call out of nowhere from 102 00:03:34,750 --> 00:03:36,706 Pfizer. And they were looking for 103 00:03:36,728 --> 00:03:38,798 somebody who had a dual 104 00:03:38,830 --> 00:03:40,942 neuroscience and genetics background. 105 00:03:41,006 --> 00:03:43,800 So it just seemed too perfect to - 106 00:03:43,800 --> 00:03:45,874 Ahead of your time, 107 00:03:45,912 --> 00:03:47,990 Chris. So when they are 108 00:03:48,060 --> 00:03:50,994 pulling those GWAS traits out of the imaging 109 00:03:51,042 --> 00:03:53,770 data, 110 00:03:53,770 --> 00:03:55,760 how is that being tracked? 111 00:03:55,760 --> 00:03:57,978 What were the 112 00:03:58,064 --> 00:03:59,898 connections you were looking for with the 113 00:03:59,904 --> 00:04:01,706 genetic data? How were 114 00:04:01,728 --> 00:04:03,770 they identified across different 115 00:04:03,840 --> 00:04:05,978 MRIs that allowed it 116 00:04:05,984 --> 00:04:07,870 to be 117 00:04:07,870 --> 00:04:09,866 compared between 118 00:04:09,968 --> 00:04:11,774 cases and controls? This 119 00:04:11,812 --> 00:04:13,310 imaging area has 120 00:04:13,380 --> 00:04:15,730 evolved so much 121 00:04:15,730 --> 00:04:17,742 since I was in graduate school, so 122 00:04:17,796 --> 00:04:19,838 I'm really curious how you did that. 123 00:04:20,004 --> 00:04:22,206 So it's interesting, I think ENIGMA was 124 00:04:22,228 --> 00:04:24,926 almost like a proto-UK 125 00:04:24,958 --> 00:04:26,574 Biobank. I think UK Biobank 126 00:04:26,622 --> 00:04:28,546 was in the middle of recruitment when 127 00:04:28,568 --> 00:04:30,834 ENIGMA started up. But really it was a great 128 00:04:30,872 --> 00:04:32,970 idea from Paul Thompson 129 00:04:32,970 --> 00:04:34,946 where there 130 00:04:34,968 --> 00:04:36,470 were a lot of different sites doing 131 00:04:36,540 --> 00:04:38,966 MRI scans in maybe 50 cases or 132 00:04:38,988 --> 00:04:40,358 50 controls, and 133 00:04:40,524 --> 00:04:42,754 reporting differences in brain 134 00:04:42,802 --> 00:04:44,806 structure and function that 135 00:04:44,908 --> 00:04:46,838 sometimes were replicated and sometimes 136 00:04:46,924 --> 00:04:48,738 weren't. So the broad sort of 137 00:04:48,764 --> 00:04:50,486 oversimplified idea of ENIGMA 138 00:04:50,518 --> 00:04:52,394 was, well, we can't bring everything 139 00:04:52,432 --> 00:04:54,586 together, we can't ask everybody to just 140 00:04:54,768 --> 00:04:56,614 throw their data in a sensor 141 00:04:56,662 --> 00:04:58,822 repository. Ethics 142 00:04:58,886 --> 00:05:00,666 and paperwork nightmare. 143 00:05:00,778 --> 00:05:02,654 But we could agree 144 00:05:02,692 --> 00:05:04,910 upon a standardized set of 145 00:05:04,980 --> 00:05:06,770 protocols 146 00:05:06,770 --> 00:05:08,954 to process the imaging 147 00:05:09,002 --> 00:05:11,774 data and ask everybody to process it 148 00:05:11,812 --> 00:05:13,666 in exactly the same way. And 149 00:05:13,688 --> 00:05:15,554 then they all send us their 150 00:05:15,592 --> 00:05:17,598 results because that's clean, it's 151 00:05:17,614 --> 00:05:19,854 anonymized, and we'll meta analyze 152 00:05:19,902 --> 00:05:21,278 all of our results together. So that's 153 00:05:21,294 --> 00:05:23,310 where the name comes from 154 00:05:23,320 --> 00:05:25,210 enhancing neuroimaging 155 00:05:25,310 --> 00:05:27,794 genetics via meta analysis. Uh, 156 00:05:27,794 --> 00:05:29,990 nice. Yeah, good memory. 157 00:05:29,990 --> 00:05:31,926 I have to 158 00:05:31,948 --> 00:05:33,462 type it out a lot during my 159 00:05:33,516 --> 00:05:35,870 post doc. That's good news, that means a lot of publications. 160 00:05:35,870 --> 00:05:37,538 161 00:05:37,564 --> 00:05:39,890 How did the 162 00:05:39,890 --> 00:05:41,866 UK Biobank come into your 163 00:05:41,888 --> 00:05:43,786 life? How did you make a connection with 164 00:05:43,808 --> 00:05:45,950 UK Biobank? And I think 165 00:05:45,950 --> 00:05:47,702 you have also seen all the progress, 166 00:05:47,766 --> 00:05:49,790 right? 167 00:05:49,790 --> 00:05:51,742 Definitely, yeah. It's interesting, 168 00:05:51,876 --> 00:05:53,806 I think that UK Biobank came into a lot 169 00:05:53,828 --> 00:05:55,500 of industry 170 00:05:55,562 --> 00:05:57,742 scientists lives around the same 171 00:05:57,796 --> 00:05:59,966 time. While I was at 172 00:05:59,988 --> 00:06:01,840 Pfizer, we were using 173 00:06:01,840 --> 00:06:03,566 large-scale 174 00:06:03,598 --> 00:06:05,720 genetic databases to 175 00:06:05,720 --> 00:06:07,618 make inferences around 176 00:06:07,784 --> 00:06:09,902 this gene is associated with this disease. 177 00:06:09,966 --> 00:06:11,534 Maybe it would make a good therapeutic 178 00:06:11,582 --> 00:06:13,454 target. But UK 179 00:06:13,502 --> 00:06:15,790 Biobank came along, I guess around 180 00:06:15,790 --> 00:06:17,466 2016, 181 00:06:17,518 --> 00:06:19,446 2017. It really started to come onto our 182 00:06:19,468 --> 00:06:21,634 radar when the exome 183 00:06:21,682 --> 00:06:23,940 sequencing of UKB was announced. 184 00:06:23,940 --> 00:06:25,814 That was one of the first sort of 185 00:06:25,852 --> 00:06:27,598 major industry- 186 00:06:27,714 --> 00:06:29,530 academic collaborations where 187 00:06:29,600 --> 00:06:31,926 UKB worked together with Pharma 188 00:06:31,958 --> 00:06:33,354 to generate the 189 00:06:33,392 --> 00:06:35,834 biggest exome sequencing study ever 190 00:06:35,872 --> 00:06:37,850 conducted. 191 00:06:37,850 --> 00:06:39,802 That came on our radar as around 192 00:06:39,856 --> 00:06:41,934 2016, 2017, I think 193 00:06:41,972 --> 00:06:43,390 for me personally, 194 00:06:43,420 --> 00:06:45,166 I moved to 195 00:06:45,188 --> 00:06:47,966 Biogen in 2018. It 196 00:06:47,988 --> 00:06:49,658 was around the time that Pfizer 197 00:06:49,834 --> 00:06:51,920 pulled out of neuroscience and 198 00:06:51,920 --> 00:06:53,646 Biogen were all in on 199 00:06:53,668 --> 00:06:54,978 neuroscience. So it just seemed like the 200 00:06:54,984 --> 00:06:56,520 perfect place for me to work. 201 00:06:56,520 --> 00:06:58,722 But the first thing I was tasked with 202 00:06:58,776 --> 00:07:00,622 when I joined Sally John's 203 00:07:00,686 --> 00:07:02,578 organization was make 204 00:07:02,664 --> 00:07:04,754 UK Biobank useful for 205 00:07:04,792 --> 00:07:06,606 neuroscience, for Multiple 206 00:07:06,638 --> 00:07:08,802 Sclerosis and Alzheimer's disease and 207 00:07:08,856 --> 00:07:10,890 depression and Parkinson's, 208 00:07:10,890 --> 00:07:12,886 et cetera. And it 209 00:07:12,908 --> 00:07:14,546 was a sort of a tall order. I mean, UK 210 00:07:14,578 --> 00:07:16,726 Biobank is breathtaking in terms of its 211 00:07:16,828 --> 00:07:18,670 depth, and 212 00:07:18,670 --> 00:07:20,218 it's just a beautiful, beautiful 213 00:07:20,304 --> 00:07:22,742 study. But it's not a disease- 214 00:07:22,806 --> 00:07:24,906 specific study. A lot of 215 00:07:24,928 --> 00:07:26,806 these diseases like Alzheimer's, 216 00:07:26,918 --> 00:07:28,954 they only come along when you hit your 217 00:07:28,992 --> 00:07:30,922 60s. So 218 00:07:30,976 --> 00:07:32,954 there's not a whole lot of people in 219 00:07:32,992 --> 00:07:34,958 there, or there weren't, at least back when 220 00:07:34,964 --> 00:07:36,574 I started working with it with 221 00:07:36,612 --> 00:07:38,654 Alzheimer's. So there was not that many 222 00:07:38,692 --> 00:07:40,762 questions that we could address using UKB. 223 00:07:40,826 --> 00:07:42,586 So the lowest hanging 224 00:07:42,618 --> 00:07:44,874 fruit for me, coming from my background 225 00:07:44,922 --> 00:07:46,578 with the ENIGMA Consortium, was to look at 226 00:07:46,584 --> 00:07:48,926 the MRI scans in UKB. Unlike 227 00:07:48,958 --> 00:07:50,926 ENIGMA, which was retrospective, 228 00:07:50,958 --> 00:07:52,894 metaanalysis, UKB 229 00:07:52,942 --> 00:07:54,750 are actually collecting 230 00:07:54,800 --> 00:07:56,562 scans across three different 231 00:07:56,616 --> 00:07:58,886 sites in the UK, all using the same type 232 00:07:58,908 --> 00:08:00,850 of scanner, the same head coil. It's 233 00:08:00,930 --> 00:08:02,980 all standardized. 234 00:08:02,980 --> 00:08:04,694 So that was the first thing I did. I did 235 00:08:04,732 --> 00:08:06,780 GWAS and a couple of new 236 00:08:06,780 --> 00:08:08,886 imaging measures from the brain 237 00:08:08,918 --> 00:08:10,520 scans. Things like 238 00:08:10,528 --> 00:08:12,830 local folding. 239 00:08:12,830 --> 00:08:14,780 But 240 00:08:14,780 --> 00:08:16,970 I felt like we could do more 241 00:08:17,040 --> 00:08:19,802 to help neuroscience. And I started 242 00:08:19,856 --> 00:08:21,726 to play around with the idea that maybe we 243 00:08:21,748 --> 00:08:23,614 could look into doing 244 00:08:23,732 --> 00:08:25,994 neurofilament light polypeptide 245 00:08:26,042 --> 00:08:28,698 or neurofilament light chain in UKB. 246 00:08:28,794 --> 00:08:30,394 So this is like a neuronal 247 00:08:30,442 --> 00:08:32,300 cytoskeletal protein. 248 00:08:32,300 --> 00:08:34,878 And when there's 249 00:08:34,884 --> 00:08:36,526 some injury, when you get neuronal 250 00:08:36,558 --> 00:08:38,706 injury, it gets 251 00:08:38,728 --> 00:08:40,846 secreted into fluids. 252 00:08:40,878 --> 00:08:42,578 So CSF [cerebrospinal fluid] or 253 00:08:42,664 --> 00:08:44,754 blood. And it 254 00:08:44,792 --> 00:08:46,390 was proposed, it was really 255 00:08:46,540 --> 00:08:48,662 gaining momentum as a 256 00:08:48,716 --> 00:08:50,540 potential biomarker for 257 00:08:50,540 --> 00:08:52,786 MS [Multiple Sclerosis] and Alzheimer's and other brain 258 00:08:52,818 --> 00:08:54,774 illnesses. So it just 259 00:08:54,812 --> 00:08:56,966 seemed like a really exciting idea. What 260 00:08:56,988 --> 00:08:58,774 if we could measure neurofilament across 261 00:08:58,892 --> 00:09:00,834 UK Biobank, across these half a million 262 00:09:00,892 --> 00:09:02,294 people, and we could get a database 263 00:09:02,342 --> 00:09:04,826 of how much brain injury do you 264 00:09:04,848 --> 00:09:06,650 have based on a blood sample? 265 00:09:06,820 --> 00:09:08,618 But quickly realized that was going to 266 00:09:08,624 --> 00:09:10,714 be very expensive and a little 267 00:09:10,752 --> 00:09:12,460 bit niche as well. 268 00:09:12,480 --> 00:09:14,800 269 00:09:14,800 --> 00:09:16,634 There's not that many 270 00:09:16,672 --> 00:09:18,238 pharmas that are invested in 271 00:09:18,244 --> 00:09:20,906 neuroscience these days. And we felt 272 00:09:20,938 --> 00:09:22,866 that if we were going to do it, we 273 00:09:22,888 --> 00:09:24,878 would need it to be a multi pharma 274 00:09:24,974 --> 00:09:26,514 consortium effort given its 275 00:09:26,552 --> 00:09:28,834 expense. So thought about 276 00:09:28,872 --> 00:09:30,950 it more and more and I 277 00:09:30,950 --> 00:09:32,980 had already worked with Olink 278 00:09:32,980 --> 00:09:34,530 on 279 00:09:34,600 --> 00:09:36,886 smaller scale studies. What year was this 280 00:09:36,908 --> 00:09:38,806 about? This was 281 00:09:38,908 --> 00:09:40,886 2018, I 282 00:09:40,908 --> 00:09:42,880 think. 2018, 2019. 283 00:09:42,880 --> 00:09:44,998 But 284 00:09:45,004 --> 00:09:46,582 I had been working with Olink 285 00:09:46,636 --> 00:09:48,694 on some smaller scale studies. I'd done some 286 00:09:48,732 --> 00:09:50,934 work in a Swedish neurology cohort 287 00:09:50,982 --> 00:09:52,714 looking into proteomic changes in 288 00:09:52,752 --> 00:09:54,906 Alzheimer's disease, and started to 289 00:09:54,928 --> 00:09:56,806 talk to Evan Mills at Olink 290 00:09:56,838 --> 00:09:58,746 about, "Hey, are you going to get 291 00:09:58,768 --> 00:10:00,886 neurofilament on Olink any day? I'd 292 00:10:00,918 --> 00:10:02,586 love to look at neurofilament in UK 293 00:10:02,618 --> 00:10:04,830 Biobank." And we started to 294 00:10:04,980 --> 00:10:06,574 toss around the idea that 295 00:10:06,692 --> 00:10:08,782 maybe, instead of just doing 296 00:10:08,836 --> 00:10:10,526 neurofilament in UKB, we could do 297 00:10:10,548 --> 00:10:12,442 Olink because it captures 298 00:10:12,506 --> 00:10:14,746 neurofilament and it captures many other 299 00:10:14,788 --> 00:10:16,354 proteins at the same time. So we 300 00:10:16,392 --> 00:10:18,930 could not just make this about 301 00:10:19,000 --> 00:10:21,886 enhancing the value for neuroscience in UK 302 00:10:21,918 --> 00:10:23,394 Biobank, but just in general, 303 00:10:23,512 --> 00:10:25,794 enhancing the value for drug discovery and 304 00:10:25,832 --> 00:10:27,686 potentially opening this up to a 305 00:10:27,708 --> 00:10:29,238 wider consortium of 306 00:10:29,324 --> 00:10:31,682 pharmas. But, yeah, that's a mouthful. 307 00:10:31,746 --> 00:10:33,900 Basically, I can't remember the question 308 00:10:33,900 --> 00:10:35,926 he asked. I asked you how 309 00:10:35,948 --> 00:10:37,794 you got started with the UK Biobank. 310 00:10:37,842 --> 00:10:39,666 And it's great because you zoomed right 311 00:10:39,708 --> 00:10:41,750 into sort of getting 13 312 00:10:41,830 --> 00:10:43,994 pharmas together. That was 313 00:10:44,032 --> 00:10:46,586 no mean feat. What was it like? I mean, here 314 00:10:46,608 --> 00:10:48,022 it is. You're going from one protein, 315 00:10:48,086 --> 00:10:50,846 realizing that NFL is not going to be 316 00:10:50,868 --> 00:10:52,814 of general interest, and then 317 00:10:52,852 --> 00:10:54,846 some exposure to Olink. There 318 00:10:54,868 --> 00:10:56,062 must have been a lot of different 319 00:10:56,116 --> 00:10:58,370 conversations. 320 00:10:58,370 --> 00:11:00,926 Yes. I don't know where to 321 00:11:00,948 --> 00:11:02,826 start. If someone walks 322 00:11:02,858 --> 00:11:04,926 up to you today and they say, how did you do 323 00:11:04,948 --> 00:11:06,946 it? How did you make that happen? What 324 00:11:06,968 --> 00:11:08,546 do you say to them? Because you shared with 325 00:11:08,568 --> 00:11:10,866 me once that was a question you 326 00:11:10,888 --> 00:11:12,870 get asked a lot. 327 00:11:12,870 --> 00:11:14,870 Yeah, 328 00:11:14,870 --> 00:11:16,882 it was a convergence 329 00:11:16,946 --> 00:11:18,886 of factors, I guess, so to 330 00:11:18,908 --> 00:11:20,998 speak. I think it was a 331 00:11:21,004 --> 00:11:23,830 mixture of it was good timing because 332 00:11:23,980 --> 00:11:25,846 I had been involved on the 333 00:11:25,868 --> 00:11:27,942 Exome Sequencing Consortium, which was 334 00:11:27,996 --> 00:11:29,830 eight different pharmaceutical companies 335 00:11:29,900 --> 00:11:31,746 funding that, and that was wrapping 336 00:11:31,778 --> 00:11:33,738 up. And we had a 337 00:11:33,744 --> 00:11:35,658 conversation amongst the eight of us of, 338 00:11:35,744 --> 00:11:37,578 what would we like to do next? Do we want to 339 00:11:37,584 --> 00:11:39,818 do something next? And we basically took a 340 00:11:39,824 --> 00:11:41,178 straw poll of other 341 00:11:41,264 --> 00:11:43,802 multiomics techniques and proteomics 342 00:11:43,866 --> 00:11:45,646 really rose to the top. So 343 00:11:45,748 --> 00:11:47,534 I saw that as an opportunity. I was a huge 344 00:11:47,572 --> 00:11:49,834 fan of proteomics to make my pitch 345 00:11:49,882 --> 00:11:51,886 to that group of 346 00:11:51,988 --> 00:11:53,626 companies. And it seemed 347 00:11:53,658 --> 00:11:55,826 to go down well, but the 348 00:11:55,848 --> 00:11:57,490 timing just happened to be right 349 00:11:57,560 --> 00:11:59,858 because the field of proteomics was 350 00:11:59,864 --> 00:12:01,982 maturing to the point where these multiplex 351 00:12:02,046 --> 00:12:04,418 technologies could capture quite a 352 00:12:04,424 --> 00:12:06,998 sizable proportion of the 353 00:12:07,084 --> 00:12:09,670 canonical human plasma proteome. 354 00:12:09,910 --> 00:12:11,750 And it just happened to be a time 355 00:12:11,820 --> 00:12:13,254 where the 356 00:12:13,292 --> 00:12:15,814 pharmas had budgets set aside to do 357 00:12:15,852 --> 00:12:17,910 something innovative like this. 358 00:12:17,910 --> 00:12:19,866 But yeah, and had a good network of 359 00:12:19,888 --> 00:12:21,814 people helping me out. Melissa Miller 360 00:12:21,862 --> 00:12:23,914 from Pfizer was a huge proponent of 361 00:12:23,952 --> 00:12:25,686 this alongside me, and Lyndon 362 00:12:25,718 --> 00:12:27,686 Mitnaul from Regeneron 363 00:12:27,798 --> 00:12:29,890 as well. So lots of different people, 364 00:12:29,890 --> 00:12:31,626 just basically all coming together and 365 00:12:31,648 --> 00:12:33,774 agreeing that this was a good idea. I have 366 00:12:33,812 --> 00:12:35,502 to just tell you that I was 367 00:12:35,556 --> 00:12:37,454 talking to someone 368 00:12:37,652 --> 00:12:39,566 on a different interview, and I said 369 00:12:39,588 --> 00:12:41,902 Melissa McCarthy, because Mark 370 00:12:41,956 --> 00:12:43,306 McCarthy and Melissa 371 00:12:43,338 --> 00:12:45,886 Miller were both involved in 372 00:12:45,908 --> 00:12:47,954 this. I just made that connection just 373 00:12:47,992 --> 00:12:49,582 now, as you said that. That's funny. 374 00:12:49,646 --> 00:12:51,906 Now, timing wise, you mentioned that you 375 00:12:51,928 --> 00:12:53,662 started talking 2018 376 00:12:53,726 --> 00:12:55,614 2019, if memory serves 377 00:12:55,662 --> 00:12:57,922 correctly. I think there was a press release 378 00:12:57,986 --> 00:12:59,878 at the end of 2020 379 00:13:00,044 --> 00:13:02,066 announcing the UK Biobank's 380 00:13:02,098 --> 00:13:04,070 involvement. So that must have been a very 381 00:13:04,140 --> 00:13:06,410 busy year and a half. 382 00:13:06,410 --> 00:13:08,806 Yeah, I've always had these bags under my 383 00:13:08,828 --> 00:13:10,960 eyes, but they got bigger 384 00:13:10,960 --> 00:13:12,750 in 2020. 385 00:13:12,750 --> 00:13:14,566 The 386 00:13:14,588 --> 00:13:16,806 first proper conversation that we 387 00:13:16,828 --> 00:13:18,294 had about this was in 388 00:13:18,412 --> 00:13:20,960 Pfizer's New York 389 00:13:21,170 --> 00:13:23,950 campus in, I think, 390 00:13:24,020 --> 00:13:26,942 May. Sorry, 391 00:13:26,996 --> 00:13:28,990 February or March, I should say, of 392 00:13:29,060 --> 00:13:31,600 2020. And I gave the pitch there. 393 00:13:31,690 --> 00:13:33,678 And yeah, 394 00:13:33,764 --> 00:13:35,374 then everything shut down. The whole world 395 00:13:35,412 --> 00:13:37,698 shut down. So, the rest of the pitch was 396 00:13:37,784 --> 00:13:39,954 virtual. So originally we got six 397 00:13:39,992 --> 00:13:41,966 of those eight companies signed 398 00:13:41,998 --> 00:13:43,938 up, and then getting the other seven on 399 00:13:43,944 --> 00:13:45,750 board was all 400 00:13:45,750 --> 00:13:47,526 meeting people for the first time from 401 00:13:47,548 --> 00:13:49,494 different pharmas that it was all 402 00:13:49,532 --> 00:13:51,870 through Zoom or 403 00:13:51,870 --> 00:13:53,846 Microsoft Teams or what have you. Do 404 00:13:53,868 --> 00:13:55,794 you think that Zoom 405 00:13:55,842 --> 00:13:57,866 was an impediment? Or do you think 406 00:13:57,888 --> 00:13:59,386 it actually because some things, 407 00:13:59,488 --> 00:14:01,790 oddly COVID, 408 00:14:01,790 --> 00:14:03,914 and this push to 409 00:14:03,952 --> 00:14:05,990 Zoom and teleconferencing 410 00:14:06,150 --> 00:14:08,886 kind of ushered in 411 00:14:08,918 --> 00:14:10,606 telehealth that probably brought us a 412 00:14:10,628 --> 00:14:12,894 decade forward in using 413 00:14:12,932 --> 00:14:14,686 telehealth solutions. I'm just 414 00:14:14,708 --> 00:14:16,398 curious about your perspective on whether you 415 00:14:16,404 --> 00:14:18,910 think that helped or hurt or was neutral. 416 00:14:18,910 --> 00:14:20,826 I have a silly perspective 417 00:14:20,858 --> 00:14:22,974 on this. I 418 00:14:23,012 --> 00:14:25,274 like it. I actually thought it was helpful 419 00:14:25,402 --> 00:14:27,426 for two very silly reasons. I think the 420 00:14:27,448 --> 00:14:29,934 first is that I can be awkward 421 00:14:29,982 --> 00:14:31,682 in person, and I'm not very good at small 422 00:14:31,736 --> 00:14:33,990 talk. So Zoom is very 423 00:14:33,990 --> 00:14:35,854 you get online and then you get straight 424 00:14:35,902 --> 00:14:37,874 into it. I've seen you in action 425 00:14:37,922 --> 00:14:39,414 and you do get straight into 426 00:14:39,452 --> 00:14:41,654 it. You get things 427 00:14:41,692 --> 00:14:43,560 done, and then 428 00:14:43,800 --> 00:14:45,942 I'm short. I'm like, five [foot] seven [inches], 429 00:14:45,996 --> 00:14:47,750 so nobody can see that on Zoom. 430 00:14:49,710 --> 00:14:51,414 Those are two very valid 431 00:14:51,462 --> 00:14:53,514 reasons. Cut out the small talk. 432 00:14:53,632 --> 00:14:55,800 And 433 00:14:55,800 --> 00:14:57,946 I just took us down a rabbit hole, but 434 00:14:57,968 --> 00:14:59,738 I love it. You 435 00:14:59,744 --> 00:15:01,860 mentioned about multiomics. 436 00:15:01,860 --> 00:15:03,834 How will you see the value of using 437 00:15:03,872 --> 00:15:05,982 multiomics in big cohorts like the 438 00:15:06,036 --> 00:15:08,126 UK Biobank? And what is the position of 439 00:15:08,148 --> 00:15:09,754 proteomics? How will you see Protonics 440 00:15:09,802 --> 00:15:11,960 position in this multiomics approach? 441 00:15:11,960 --> 00:15:13,966 Yeah, that's a good question. And 442 00:15:14,148 --> 00:15:15,854 sorry if this sounds a little 443 00:15:15,892 --> 00:15:17,842 rehearsed. It isn't. But I've given so many 444 00:15:17,896 --> 00:15:19,406 talks on this at this point that I'll 445 00:15:19,438 --> 00:15:21,922 probably say the same thing that I often do, 446 00:15:21,976 --> 00:15:23,458 which is that we've been using 447 00:15:23,624 --> 00:15:25,614 UK Biobank and FinnGen and 448 00:15:25,672 --> 00:15:27,874 these big population biobanks 449 00:15:27,922 --> 00:15:29,494 to make links between 450 00:15:29,692 --> 00:15:31,814 gene variants and diseases, and then 451 00:15:31,852 --> 00:15:33,622 turn those links into 452 00:15:33,756 --> 00:15:35,942 new drugs. So gene "X" is a really strong 453 00:15:35,996 --> 00:15:37,762 association with disease 454 00:15:37,826 --> 00:15:39,982 "Y". Let's turn it into a new drug. 455 00:15:40,066 --> 00:15:42,054 Let's make a small molecule or an antibody 456 00:15:42,102 --> 00:15:44,590 that hits the protein 457 00:15:44,590 --> 00:15:46,778 that's encoded by that gene. Now 458 00:15:46,864 --> 00:15:48,426 that hits the protein. We're not 459 00:15:48,448 --> 00:15:50,860 measuring the protein, and that's the issue. 460 00:15:50,900 --> 00:15:52,930 We're doing GWAS, 461 00:15:52,950 --> 00:15:54,990 we're finding lots of new genes, 462 00:15:55,060 --> 00:15:57,838 and a lot of them are intriguing, but a 463 00:15:57,844 --> 00:15:59,566 lot of them are very difficult to drug. And 464 00:15:59,588 --> 00:16:01,650 a lot of the time, 465 00:16:01,650 --> 00:16:03,806 the gene association that we've 466 00:16:03,838 --> 00:16:05,822 identified, it's messy. 467 00:16:05,886 --> 00:16:07,886 I mean, it takes a lot of downstream 468 00:16:07,918 --> 00:16:09,986 work to pinpoint exactly what gene it 469 00:16:10,008 --> 00:16:12,890 is. And oftentimes, 470 00:16:12,890 --> 00:16:14,930 it's either not completely clear 471 00:16:15,000 --> 00:16:17,734 or it's very pleiotropic, where it could be 472 00:16:17,772 --> 00:16:19,830 affecting a lot of different proteins 473 00:16:19,830 --> 00:16:21,590 or pathways. So, 474 00:16:21,660 --> 00:16:23,846 really, I always thought proteins as the 475 00:16:23,868 --> 00:16:25,530 missing piece between 476 00:16:25,530 --> 00:16:27,814 genes and diseases in that 477 00:16:27,852 --> 00:16:29,830 genetics guided drug discovery process. 478 00:16:29,980 --> 00:16:31,926 The proteins, we could argue 479 00:16:31,958 --> 00:16:33,606 about it about how much they represent drug 480 00:16:33,638 --> 00:16:35,466 targets now that we have gene therapies and 481 00:16:35,488 --> 00:16:37,782 siRNAs, et cetera, that don't necessarily 482 00:16:37,846 --> 00:16:39,610 target proteins, but 483 00:16:39,680 --> 00:16:41,862 still, especially for bigger pharma, 484 00:16:41,926 --> 00:16:43,578 the vast majority of the drugs we're 485 00:16:43,594 --> 00:16:45,822 making are targeting proteins. So let's put 486 00:16:45,876 --> 00:16:47,946 our drug targets part and parcel 487 00:16:47,978 --> 00:16:49,626 of that genetic drug discovery 488 00:16:49,658 --> 00:16:51,966 process, and then we have the potential to 489 00:16:51,988 --> 00:16:53,874 maybe reveal something mechanistic about 490 00:16:53,912 --> 00:16:55,860 how that drug is acting as well. 491 00:16:55,860 --> 00:16:57,794 Exactly. 492 00:16:57,912 --> 00:16:59,780 Yeah. And from the 493 00:16:59,780 --> 00:17:01,806 pharmaceutical 494 00:17:01,806 --> 00:17:03,490 drug discovery angle, 495 00:17:03,520 --> 00:17:05,846 they intuitively sort of picked this 496 00:17:05,868 --> 00:17:07,494 up, meaning they accepted that 497 00:17:07,532 --> 00:17:09,414 premise that we go from 498 00:17:09,452 --> 00:17:11,798 genetic guided drug discovery to 499 00:17:11,884 --> 00:17:13,542 gene, to protein, to 500 00:17:13,596 --> 00:17:15,810 disease. 501 00:17:15,810 --> 00:17:17,938 I hope that they liked 502 00:17:17,954 --> 00:17:19,686 it. They seem to like it because they 503 00:17:19,708 --> 00:17:21,706 invested in the PPP [Pharma Proteomics] project. But, 504 00:17:21,728 --> 00:17:23,800 yeah, I think that 505 00:17:23,800 --> 00:17:25,930 it wasn't a difficult 506 00:17:26,000 --> 00:17:27,802 argument to make, because I think people 507 00:17:27,856 --> 00:17:29,526 have seen there've been a couple of papers 508 00:17:29,558 --> 00:17:31,918 from AstraZeneca and Abbvie and 509 00:17:31,924 --> 00:17:33,914 others, and they've looked retrospectively 510 00:17:33,962 --> 00:17:35,802 at their drug development pipelines. 511 00:17:35,866 --> 00:17:37,674 And they've basically assessed, 512 00:17:37,722 --> 00:17:39,982 okay, which drugs made it 513 00:17:40,116 --> 00:17:42,526 to patients and which drugs failed, and then 514 00:17:42,628 --> 00:17:44,806 which drugs had support from GWAS 515 00:17:44,858 --> 00:17:46,798 or ClinVar association 516 00:17:46,894 --> 00:17:48,786 and which ones didn't. And there have been a 517 00:17:48,808 --> 00:17:50,114 couple of independent studies that have 518 00:17:50,152 --> 00:17:52,860 shown that if your drug target has 519 00:17:53,016 --> 00:17:54,958 supporting evidence from genetics, then it's 520 00:17:54,974 --> 00:17:56,802 at least twice as likely to actually 521 00:17:56,856 --> 00:17:58,978 succeed. But there's a 522 00:17:58,984 --> 00:18:00,674 lot of unanswered questions there 523 00:18:00,792 --> 00:18:02,534 that seems to be pretty good evidence. Yeah, 524 00:18:02,572 --> 00:18:04,306 okay, let's use genetics for drug discovery, 525 00:18:04,338 --> 00:18:06,518 but there's a lot of murky stuff in 526 00:18:06,524 --> 00:18:08,826 the middle that we still need to figure out. 527 00:18:08,848 --> 00:18:10,746 So I think that's where the multiomics can 528 00:18:10,768 --> 00:18:12,986 help. And as far 529 00:18:13,008 --> 00:18:15,594 as the Pharma Proteomics Project 530 00:18:15,712 --> 00:18:17,930 being, frankly, 531 00:18:17,930 --> 00:18:19,994 you can say 532 00:18:20,032 --> 00:18:21,706 it's a pilot, right? Because you're looking 533 00:18:21,728 --> 00:18:23,378 at one-tength the size of the UK 534 00:18:23,414 --> 00:18:25,614 Biobank. You can also make the 535 00:18:25,652 --> 00:18:27,806 argument that, well, something like this has 536 00:18:27,828 --> 00:18:29,886 not been done at this scale before in terms 537 00:18:29,908 --> 00:18:31,950 of looking at 1500 proteins. 538 00:18:31,950 --> 00:18:33,662 Were you 539 00:18:33,716 --> 00:18:35,966 pointing to other work that had looked at 540 00:18:35,988 --> 00:18:37,970 circulating proteins in genetics 541 00:18:37,970 --> 00:18:39,966 as far as mendelian randomization, 542 00:18:40,078 --> 00:18:42,594 that kind of thing? Yeah, absolutely. 543 00:18:42,712 --> 00:18:44,574 There's been a couple of big studies. 544 00:18:44,622 --> 00:18:46,750 Claudia Langenberg is one of the pioneers in 545 00:18:46,760 --> 00:18:48,410 this field. She's awesome. 546 00:18:48,410 --> 00:18:50,946 Well, I didn't prepare for this. I'm worried 547 00:18:50,978 --> 00:18:52,178 I'm going to leave people out. But there's 548 00:18:52,194 --> 00:18:54,418 Claudia, of course. There's Kári 549 00:18:54,434 --> 00:18:56,326 Stefánsson in Iceland with 550 00:18:56,348 --> 00:18:58,918 deCODE [Genetics]. Yeah, several different ... 551 00:18:59,004 --> 00:19:00,738 There's the SCALLOP consortium that we're 552 00:19:00,754 --> 00:19:02,758 doing this at a, I won't say smaller scale 553 00:19:02,774 --> 00:19:04,378 because they'd amassed quite a large 554 00:19:04,464 --> 00:19:06,874 collection of Olink data, but 555 00:19:06,912 --> 00:19:08,986 just based on the old panels. So kind of 556 00:19:09,008 --> 00:19:11,578 90 proteins at a time. So there had been a 557 00:19:11,584 --> 00:19:13,422 lot, a lot of precedence. This definitely 558 00:19:13,476 --> 00:19:15,374 wasn't the first time anybody was doing 559 00:19:15,412 --> 00:19:17,250 this. It just happened to be the biggest 560 00:19:17,250 --> 00:19:19,902 so far. So their 561 00:19:19,956 --> 00:19:21,806 appetite was whetted. In 562 00:19:21,828 --> 00:19:23,890 terms of these smaller studies, 563 00:19:23,890 --> 00:19:25,966 they knew that this approach could work 564 00:19:25,988 --> 00:19:27,694 and therefore that was really a low risk 565 00:19:27,742 --> 00:19:29,506 decision. Do I understand that 566 00:19:29,528 --> 00:19:31,854 correctly? To a certain degree. 567 00:19:31,902 --> 00:19:33,106 I think that there were two ways you could 568 00:19:33,128 --> 00:19:34,738 have pitched this. You could have pitched it 569 00:19:34,744 --> 00:19:36,470 to 570 00:19:36,470 --> 00:19:38,390 geneticists or you could have pitched it to 571 00:19:38,460 --> 00:19:40,662 biomarker experts or 572 00:19:40,716 --> 00:19:42,790 proteomics experts. And 573 00:19:42,940 --> 00:19:44,838 I felt that the pitch was easier to the 574 00:19:44,844 --> 00:19:46,886 geneticist because genetics for at least the 575 00:19:46,908 --> 00:19:48,694 last 15 years, if not 576 00:19:48,732 --> 00:19:50,966 longer, is used to doing things at 577 00:19:50,988 --> 00:19:52,750 very large scale. 578 00:19:52,750 --> 00:19:54,794 You need to do things in tens and now 579 00:19:54,832 --> 00:19:56,854 hundreds of thousands. And some of the GWASes 580 00:19:56,902 --> 00:19:58,986 are now even in over a million now in 581 00:19:59,008 --> 00:20:00,906 order to pick up the biology, in order to 582 00:20:00,928 --> 00:20:02,426 pick up the gene variants that are 583 00:20:02,448 --> 00:20:04,922 influencing your disease. So they're used 584 00:20:04,976 --> 00:20:06,798 to doing things at really large scale. I 585 00:20:06,804 --> 00:20:08,266 think that they don't need to be convinced 586 00:20:08,298 --> 00:20:10,158 of that. I think the biomarker folks are 587 00:20:10,164 --> 00:20:12,718 more about let's do it with precision. I 588 00:20:12,724 --> 00:20:14,326 think that they still needed some convincing 589 00:20:14,378 --> 00:20:16,174 that we could do this at massive, massive 590 00:20:16,222 --> 00:20:18,894 scale. But do you think the NGS [next-generation sequencing] approach 591 00:20:18,942 --> 00:20:20,962 help you to make your 592 00:20:21,016 --> 00:20:23,666 pitch to the geneticists because it's an 593 00:20:23,688 --> 00:20:25,954 NGS approach and maybe they are more 594 00:20:25,992 --> 00:20:27,926 familiar with this approach? How was your 595 00:20:27,948 --> 00:20:29,846 feeling? Yeah, I 596 00:20:29,868 --> 00:20:31,814 think so. I think a lot of folks had 597 00:20:31,852 --> 00:20:33,942 used Olink before, 598 00:20:33,996 --> 00:20:35,958 I think using the prior sort of 599 00:20:35,964 --> 00:20:37,990 method, the PCR-based method. 600 00:20:38,130 --> 00:20:40,666 I think that we'd seen some good 601 00:20:40,688 --> 00:20:42,938 quality based on those data and felt that 602 00:20:43,104 --> 00:20:45,654 the jump to NGS would allow us to scale 603 00:20:45,702 --> 00:20:47,860 up like this. 604 00:20:51,970 --> 00:20:53,386 What is the next step from the UK 605 00:20:53,418 --> 00:20:55,966 Biobank? What's your 606 00:20:55,988 --> 00:20:57,566 ambition actually first, and what's the next 607 00:20:57,588 --> 00:20:59,980 step of a UK Biobank? 608 00:20:59,980 --> 00:21:01,870 Yeah, 609 00:21:01,870 --> 00:21:03,934 obviously it would be great to do all 610 00:21:03,972 --> 00:21:05,326 half a million. And I think that we're 611 00:21:05,358 --> 00:21:07,090 talking about that. We're having early 612 00:21:07,160 --> 00:21:09,090 conversations about whether that will be 613 00:21:09,240 --> 00:21:11,778 feasible financially more than 614 00:21:11,864 --> 00:21:13,326 sort of technically. I think that it's 615 00:21:13,358 --> 00:21:15,586 starting to become technically possible, but 616 00:21:15,688 --> 00:21:17,278 we have to talk about how much it would 617 00:21:17,304 --> 00:21:19,750 cost. I think in the shorter term, 618 00:21:19,750 --> 00:21:21,926 we're hoping, and I hope I don't 619 00:21:21,948 --> 00:21:23,686 jinx it by announcing it here, but we 620 00:21:23,708 --> 00:21:25,826 have received approval 621 00:21:25,858 --> 00:21:27,782 to do a smaller follow-up 622 00:21:27,836 --> 00:21:29,670 study in 2500 623 00:21:29,740 --> 00:21:31,738 samples in the UK Biobank. And 624 00:21:31,824 --> 00:21:33,626 these 2500 samples have 625 00:21:33,728 --> 00:21:35,878 already been profiled using the Olink 626 00:21:35,894 --> 00:21:37,498 Explore assay. But we're going to 627 00:21:37,584 --> 00:21:39,562 do three mass spec-based 628 00:21:39,616 --> 00:21:41,226 platforms from Seer and 629 00:21:41,248 --> 00:21:43,710 Biognosys and Eliptica, as well as 630 00:21:43,710 --> 00:21:45,950 SomaLogic and then we'll just have a 631 00:21:46,100 --> 00:21:48,930 very comprehensive 632 00:21:48,930 --> 00:21:50,618 characterization of the plasma 633 00:21:50,634 --> 00:21:52,926 proteome in these 2500 people. And some 634 00:21:52,948 --> 00:21:54,814 of these people would have had COVID before 635 00:21:54,852 --> 00:21:56,338 they entered the study and some didn't. So 636 00:21:56,344 --> 00:21:58,830 it's sort of like, let's try to 637 00:21:58,830 --> 00:22:00,814 capture as much of the plasma proteome 638 00:22:00,862 --> 00:22:02,910 pre- and post-COVID as we can. 639 00:22:02,910 --> 00:22:04,674 That'll be so 640 00:22:04,712 --> 00:22:06,886 interesting, I think, especially to see 641 00:22:06,908 --> 00:22:08,294 how the complementarity of these 642 00:22:08,332 --> 00:22:10,850 technologies 643 00:22:10,970 --> 00:22:12,966 wins out in a big cohort like 644 00:22:12,988 --> 00:22:14,818 this. What are you able to reveal 645 00:22:14,914 --> 00:22:16,810 if Seer has 646 00:22:16,810 --> 00:22:18,874 this vision to be able 647 00:22:18,912 --> 00:22:20,906 to sequence the proteome, try 648 00:22:20,928 --> 00:22:22,534 and look at things that aren't 649 00:22:22,582 --> 00:22:24,960 targeted, whereas some of the others, 650 00:22:24,960 --> 00:22:26,954 are - we go 651 00:22:26,992 --> 00:22:28,986 after targeted proteins. And 652 00:22:29,008 --> 00:22:30,774 then I think these mass spec technologies 653 00:22:30,822 --> 00:22:32,794 are well established as gold 654 00:22:32,842 --> 00:22:34,954 standards and have advanced 655 00:22:35,082 --> 00:22:37,706 very far in the last few years in throughput. 656 00:22:37,898 --> 00:22:39,390 Because you're looking at different 657 00:22:39,460 --> 00:22:41,646 things, right? In terms of what kind of 658 00:22:41,748 --> 00:22:43,822 overlap there is with the canonical 659 00:22:43,886 --> 00:22:45,906 protein or versus sort 660 00:22:45,928 --> 00:22:47,746 of splice isoforms and all the 661 00:22:47,768 --> 00:22:49,886 variety of proteoforms. I mean, oh my gosh, 662 00:22:49,918 --> 00:22:51,698 there's what, 400 different 663 00:22:51,784 --> 00:22:53,614 types of post translational 664 00:22:53,662 --> 00:22:55,750 modifications. I mean you can 665 00:22:55,750 --> 00:22:57,880 just start multiplying numbers together. 666 00:22:57,880 --> 00:22:59,782 When people 667 00:22:59,836 --> 00:23:01,462 ask you, because they've asked me this, 668 00:23:01,516 --> 00:23:03,622 Chris, how many proteins do you think 669 00:23:03,676 --> 00:23:05,234 are there, including 670 00:23:05,282 --> 00:23:07,622 proteoforms, what is your thought about 671 00:23:07,676 --> 00:23:09,870 that? You can't 672 00:23:09,890 --> 00:23:11,834 have a wrong answer because we 673 00:23:11,872 --> 00:23:13,950 don't know. 674 00:23:13,950 --> 00:23:15,706 It gets kind of mind 675 00:23:15,728 --> 00:23:17,290 boggling to think about, because obviously, 676 00:23:17,440 --> 00:23:19,338 without the proteoforms, you would expect 677 00:23:19,424 --> 00:23:21,434 there to be 20,000 just based on the human 678 00:23:21,472 --> 00:23:23,834 genome. But then it depends on how many 679 00:23:23,872 --> 00:23:25,966 you can capture in blood. In terms 680 00:23:25,988 --> 00:23:27,246 of how they're expressed in different 681 00:23:27,268 --> 00:23:29,310 tissues, proteoforms, I don't know. 682 00:23:29,380 --> 00:23:30,894 Whatever I say will probably make me sound 683 00:23:30,932 --> 00:23:32,478 dumb. Especially in five or ten years when 684 00:23:32,484 --> 00:23:34,670 they work it out, like 100,000, 685 00:23:34,740 --> 00:23:36,898 maybe. Yeah, that's what I've said. 686 00:23:37,064 --> 00:23:39,074 I think I read somewhere someone made a good 687 00:23:39,112 --> 00:23:41,166 argument around that. Maybe it was Karsten [Suhre], 688 00:23:41,198 --> 00:23:43,730 maybe it was Jochen [Schwenk]. I don't know. Someone 689 00:23:43,800 --> 00:23:45,934 said that. It reminds me of the speculation 690 00:23:45,982 --> 00:23:47,970 of how many genes were in the 691 00:23:47,970 --> 00:23:49,874 Genome Project. 692 00:23:49,912 --> 00:23:51,622 The numbers were all over the place. 693 00:23:51,756 --> 00:23:53,398 I mean, who would have guessed it would be a 694 00:23:53,404 --> 00:23:55,622 little bit less than 20,000? I mean, not 695 00:23:55,676 --> 00:23:57,334 that many, right? A lot of people really 696 00:23:57,372 --> 00:23:59,466 thought it was a much, much larger number. 697 00:23:59,648 --> 00:24:01,898 100%. Yeah, exactly. 698 00:24:02,064 --> 00:24:04,938 And then, as far as I understand 699 00:24:05,024 --> 00:24:07,738 that impending - or I'm sorry, already 700 00:24:07,824 --> 00:24:09,914 we've got released the data in terms 701 00:24:09,952 --> 00:24:11,906 of the Olink first 702 00:24:11,908 --> 00:24:13,614 1500 [participants in the UK Biobank] against the 703 00:24:13,652 --> 00:24:15,490 50,000? 704 00:24:15,490 --> 00:24:17,934 Yes. I think that they are on 705 00:24:17,972 --> 00:24:19,914 the Research Access portal. 706 00:24:19,962 --> 00:24:21,946 Now, don't quote me on that. I do not represent 707 00:24:21,978 --> 00:24:23,630 UK Biobank, but I think that they are. 708 00:24:23,700 --> 00:24:25,986 Naomi told me Monday, no told me 709 00:24:26,008 --> 00:24:28,882 Friday that she said it was up there. 710 00:24:28,936 --> 00:24:30,946 So by the time this podcast comes out, 711 00:24:30,968 --> 00:24:32,994 I think you're pretty safe. 712 00:24:33,032 --> 00:24:35,830 There's several weeks-long lag time here, 713 00:24:35,900 --> 00:24:37,798 so we're looking at May 714 00:24:37,884 --> 00:24:39,900 2023, 715 00:24:39,900 --> 00:24:41,542 the first sort of set 716 00:24:41,596 --> 00:24:43,222 of roughly how many 717 00:24:43,276 --> 00:24:45,606 samples? It's probably about 718 00:24:45,708 --> 00:24:47,942 54, maybe 52 after 719 00:24:47,996 --> 00:24:49,980 QC. A thousand 720 00:24:49,980 --> 00:24:51,478 samples. 52,000 721 00:24:51,564 --> 00:24:52,950 samples times some 722 00:24:53,020 --> 00:24:55,946 1469 or 723 00:24:55,968 --> 00:24:57,520 so, give or take, 724 00:24:57,520 --> 00:24:59,926 proteins analyzed by Explore 725 00:24:59,958 --> 00:25:01,614 1536. I mean, 726 00:25:01,652 --> 00:25:03,710 that's quite a data set 727 00:25:03,780 --> 00:25:05,840 for people to dig into. 728 00:25:05,840 --> 00:25:07,646 I think - go ahead. 729 00:25:07,748 --> 00:25:09,614 No. Go ahead, Dale. Sorry. I was thinking 730 00:25:09,652 --> 00:25:11,134 about all the posters at 731 00:25:11,172 --> 00:25:13,886 ASHG [American Society of Human Genetics conference], right, in October that we were 732 00:25:13,908 --> 00:25:15,934 talking about on the podcast, as far as how 733 00:25:15,972 --> 00:25:17,746 many there was, what, 19 or 734 00:25:17,768 --> 00:25:19,998 so abstracts of different types 735 00:25:20,014 --> 00:25:22,898 of work. Yeah. This is not bragging. I 736 00:25:22,904 --> 00:25:24,334 have to keep track of this so I can convince 737 00:25:24,382 --> 00:25:26,306 people at Janssen and other companies that 738 00:25:26,328 --> 00:25:28,834 this is a return on investment. But yeah, 739 00:25:29,032 --> 00:25:31,410 you should brag. I think it was 19 740 00:25:31,490 --> 00:25:33,798 abstracts and six talks at 741 00:25:33,804 --> 00:25:35,798 ASHG. But what I'm really excited about the 742 00:25:35,804 --> 00:25:36,726 public release is 743 00:25:36,748 --> 00:25:38,966 that's obviously a lot of 744 00:25:38,988 --> 00:25:40,786 output, especially for a data 745 00:25:40,828 --> 00:25:42,830 set that's so new, but 746 00:25:42,830 --> 00:25:44,918 I don't even feel like that's 747 00:25:44,934 --> 00:25:46,298 not scratching the surface even. I think 748 00:25:46,304 --> 00:25:48,186 there's going to be so much more that 749 00:25:48,208 --> 00:25:50,646 academics can do. There's a lot of creative 750 00:25:50,678 --> 00:25:51,958 things that you could do with this data set 751 00:25:51,984 --> 00:25:53,930 that might not have immediate translational 752 00:25:54,010 --> 00:25:56,540 impact for drug discovery, but academic 753 00:25:56,540 --> 00:25:58,206 scientists are going to take this and 754 00:25:58,228 --> 00:26:00,506 probably do something really revolutionary 755 00:26:00,618 --> 00:26:02,894 with it. I can't wait till next 756 00:26:02,932 --> 00:26:04,574 year's ASHG once these 757 00:26:04,612 --> 00:26:06,098 publications start getting into the 758 00:26:06,104 --> 00:26:08,466 literature, right? It's going to be all over 759 00:26:08,488 --> 00:26:10,802 the place again. Is that because 760 00:26:10,856 --> 00:26:12,434 there's such a wide variety of 761 00:26:12,472 --> 00:26:14,546 different phenotypes that they can 762 00:26:14,568 --> 00:26:16,980 associate protein level and genetics to? 763 00:26:16,980 --> 00:26:18,870 Yes, 764 00:26:18,870 --> 00:26:20,562 well, yes, to a certain degree. I think 765 00:26:20,616 --> 00:26:22,166 we've looked at that. I think probably a lot 766 00:26:22,188 --> 00:26:23,378 of the companies have looked at that. We've 767 00:26:23,394 --> 00:26:25,894 done kind of an all by all. Take all of the 768 00:26:25,932 --> 00:26:27,958 ICD [International Classification of Diseases] codes or the feed codes, and 769 00:26:27,964 --> 00:26:29,882 then run a regression against 770 00:26:29,936 --> 00:26:31,834 all the proteins. And that 771 00:26:31,872 --> 00:26:33,766 basically gives you a crude biomarker 772 00:26:33,798 --> 00:26:35,926 study. And we've been using those results 773 00:26:35,958 --> 00:26:37,820 in house. But 774 00:26:37,820 --> 00:26:39,754 I'm mainly thinking just about 775 00:26:39,792 --> 00:26:41,974 how there's so much creativity 776 00:26:42,022 --> 00:26:43,674 out there in the academic community. There's 777 00:26:43,722 --> 00:26:45,134 questions that you could address with these 778 00:26:45,172 --> 00:26:46,766 data that we probably haven't even thought 779 00:26:46,788 --> 00:26:48,730 of yet, because this was 780 00:26:48,730 --> 00:26:50,798 UK-PPP was like, one project 781 00:26:50,884 --> 00:26:52,978 out of several on our plates and 782 00:26:52,984 --> 00:26:54,530 pharma. So I think that 783 00:26:54,600 --> 00:26:56,718 fingers crossed, the academic 784 00:26:56,734 --> 00:26:58,242 community will have a lot of fun 785 00:26:58,296 --> 00:27:00,754 ideas. Well, we 786 00:27:00,792 --> 00:27:02,674 had pushed out 787 00:27:02,712 --> 00:27:04,882 an Explore 1536 data 788 00:27:04,936 --> 00:27:06,962 set when we first launched that Explore 789 00:27:07,026 --> 00:27:09,782 platform, and it was on COVID. And 790 00:27:09,836 --> 00:27:11,814 there have been publications spurred from 791 00:27:11,852 --> 00:27:13,894 that by just comparing those 792 00:27:13,932 --> 00:27:15,734 COVID data in that 793 00:27:15,772 --> 00:27:17,570 cohort over to 794 00:27:17,740 --> 00:27:19,834 whatever work the researcher was 795 00:27:19,872 --> 00:27:21,834 already doing to look at those different 796 00:27:21,872 --> 00:27:23,630 signatures. So 797 00:27:23,630 --> 00:27:25,450 seeing publicly available data 798 00:27:25,520 --> 00:27:27,850 spur novel comparisons 799 00:27:27,850 --> 00:27:29,866 and novel publications. I 800 00:27:29,888 --> 00:27:31,886 just think that's what 801 00:27:31,908 --> 00:27:33,630 it's all about, right? Getting these 802 00:27:33,700 --> 00:27:35,902 creative minds on it, crowdsourcing these 803 00:27:35,956 --> 00:27:37,550 ideas and how people 804 00:27:37,620 --> 00:27:39,834 debate ways to do things on Twitter, 805 00:27:39,882 --> 00:27:41,770 I just absolutely love. 806 00:27:41,770 --> 00:27:43,998 It's fantastic. On 807 00:27:44,004 --> 00:27:46,850 UKB, I think it kicked off in 2006, 808 00:27:46,920 --> 00:27:48,834 so it's not a new 809 00:27:48,872 --> 00:27:50,638 study, but it always feels new. They're 810 00:27:50,654 --> 00:27:52,862 always adding cool, innovative 811 00:27:52,926 --> 00:27:54,962 new technologies to this data 812 00:27:55,016 --> 00:27:57,778 set. So it'll go on for a long time to 813 00:27:57,784 --> 00:27:59,926 come, for sure. And then as far 814 00:27:59,948 --> 00:28:01,926 as you being how, 815 00:28:01,948 --> 00:28:03,846 do I say that, that organizer. You were 816 00:28:03,868 --> 00:28:05,734 there at the beginning, you must have 817 00:28:05,772 --> 00:28:07,994 lots and lots of invitations to give these 818 00:28:08,032 --> 00:28:10,774 kinds of talks. As far as UK 819 00:28:10,822 --> 00:28:12,586 Biobank and the PPP in 820 00:28:12,608 --> 00:28:14,590 particular. 821 00:28:14,590 --> 00:28:16,954 Yeah, I do. Don't ask 822 00:28:16,992 --> 00:28:18,246 why did he accept 823 00:28:18,278 --> 00:28:20,800 ours? 824 00:28:22,800 --> 00:28:24,710 No, 825 00:28:24,800 --> 00:28:26,606 I do. Yeah, it's exciting, and I want to 826 00:28:26,628 --> 00:28:28,686 make sure that I spread it around. I guess I 827 00:28:28,708 --> 00:28:30,510 am the P.I. for the study 828 00:28:30,580 --> 00:28:32,926 overall, but there's no way this ever 829 00:28:32,948 --> 00:28:34,990 would have happened without a lot of other, 830 00:28:34,990 --> 00:28:36,914 more talented and more 831 00:28:36,952 --> 00:28:38,654 intelligent people than me involved. 832 00:28:38,702 --> 00:28:40,962 So, I get invited a lot, but 833 00:28:41,096 --> 00:28:43,730 when I can to try to forward along to 834 00:28:43,730 --> 00:28:45,846 other folks who help build this 835 00:28:45,868 --> 00:28:47,714 as well: Melissa [Miller], Ben 836 00:28:47,762 --> 00:28:49,430 Sun, 837 00:28:49,430 --> 00:28:51,490 Joseph Stakowski, 838 00:28:51,490 --> 00:28:53,922 and by the way, Brad Gibson 839 00:28:53,986 --> 00:28:55,850 from Amgen 840 00:28:55,858 --> 00:28:57,586 was a huge proponent because 841 00:28:57,708 --> 00:28:59,690 the consortium was 842 00:28:59,840 --> 00:29:01,210 90 something percent 843 00:29:01,360 --> 00:29:03,466 geneticists. Brad is the 844 00:29:03,488 --> 00:29:05,610 proteomics expert. Brad has the real 845 00:29:05,680 --> 00:29:07,870 background, the hardcore background 846 00:29:07,870 --> 00:29:09,994 mass spec. So he helped put 847 00:29:10,032 --> 00:29:11,898 guardrails on this and make sure that we 848 00:29:11,904 --> 00:29:13,798 were doing things properly. Make sure 849 00:29:13,824 --> 00:29:15,854 he got the ball over 850 00:29:15,892 --> 00:29:17,822 the finish line, too, right? In terms 851 00:29:17,876 --> 00:29:19,742 of that extra weight of 852 00:29:19,796 --> 00:29:21,866 somebody who is not coming from the genetics 853 00:29:21,898 --> 00:29:23,886 field, but within the pharma proteomics 854 00:29:23,998 --> 00:29:25,990 sort of context. 855 00:29:25,990 --> 00:29:27,810 I think so. 856 00:29:27,810 --> 00:29:29,346 I have 857 00:29:29,528 --> 00:29:31,750 probably 858 00:29:31,750 --> 00:29:33,906 built a little bit of a reputation in this 859 00:29:33,928 --> 00:29:35,860 study, but I didn't really have any 860 00:29:35,860 --> 00:29:37,922 before I started it. And 861 00:29:37,976 --> 00:29:39,766 I think when I was pitching this idea, 862 00:29:39,868 --> 00:29:41,362 there was probably a lot of skepticism, 863 00:29:41,426 --> 00:29:43,286 like who's this little twerp? And he has 864 00:29:43,308 --> 00:29:45,794 his genetics background. So Brad 865 00:29:45,842 --> 00:29:47,846 being on board and putting his 866 00:29:47,868 --> 00:29:49,866 weight behind it. Mark McCarthy, as you 867 00:29:49,888 --> 00:29:51,626 mentioned earlier, Cindy is involved as 868 00:29:51,648 --> 00:29:53,946 well. And Carrie, there were a lot of 869 00:29:53,968 --> 00:29:55,546 people that 870 00:29:55,648 --> 00:29:57,962 are more prestigious than me, 871 00:29:58,016 --> 00:30:00,670 put their weight behind it, and really 872 00:30:00,670 --> 00:30:02,842 helped put it over the finish line. Well, 873 00:30:02,896 --> 00:30:04,734 they got behind your vision. That's got to 874 00:30:04,772 --> 00:30:06,850 feel good. 875 00:30:06,850 --> 00:30:08,942 But how will you see - I have a question 876 00:30:08,996 --> 00:30:10,926 now, generally more for the cohorts. How 877 00:30:10,948 --> 00:30:12,762 will you see the use of cohorts 878 00:30:12,906 --> 00:30:14,794 in pharma and the drug development 879 00:30:14,842 --> 00:30:16,706 process? I mean, what is the value and 880 00:30:16,808 --> 00:30:18,978 what do you think is having different, 881 00:30:19,064 --> 00:30:21,370 also ethnicities and different, 882 00:30:21,370 --> 00:30:23,474 let's say, from different 883 00:30:23,592 --> 00:30:25,990 places the cohorts will help on that? 884 00:30:26,060 --> 00:30:28,422 What is your vision? How's your idea 885 00:30:28,476 --> 00:30:30,810 about that? 886 00:30:30,810 --> 00:30:32,738 I think that they're 887 00:30:32,834 --> 00:30:34,770 the engine for 888 00:30:34,770 --> 00:30:35,878 sort of 889 00:30:35,884 --> 00:30:37,630 epidemiological 890 00:30:37,630 --> 00:30:39,990 health studies, basically 891 00:30:39,990 --> 00:30:41,690 for any sort of common 892 00:30:41,760 --> 00:30:43,706 complex disease, we need these 893 00:30:43,808 --> 00:30:45,626 population cohorts to 894 00:30:45,728 --> 00:30:47,626 gain a better understanding of their 895 00:30:47,648 --> 00:30:49,926 molecular mechanisms, the causal 896 00:30:49,958 --> 00:30:51,990 mechanisms, as well as potentially some of 897 00:30:52,000 --> 00:30:54,174 the environmental influences on these 898 00:30:54,212 --> 00:30:56,858 diseases. So, I think that we've 899 00:30:56,874 --> 00:30:58,714 done a lot with UKB. There's a huge push 900 00:30:58,762 --> 00:31:00,666 now, a very well deserved 901 00:31:00,698 --> 00:31:02,458 push towards looking into 902 00:31:02,564 --> 00:31:04,910 underrepresented population cohorts. 903 00:31:04,990 --> 00:31:06,898 So lots of different ones that we 904 00:31:06,904 --> 00:31:08,750 could potentially look into. 905 00:31:08,750 --> 00:31:10,950 And also 906 00:31:10,950 --> 00:31:12,846 disease enriched cohorts, cohorts 907 00:31:12,878 --> 00:31:14,590 that might have a dementia 908 00:31:14,670 --> 00:31:16,930 wing, for example. I know that Finngen 909 00:31:17,010 --> 00:31:19,842 is building up its dementia substudy, 910 00:31:19,906 --> 00:31:21,478 so lots of different 911 00:31:21,564 --> 00:31:23,942 directions that could go in. Is there a 912 00:31:23,996 --> 00:31:25,698 threshold, a minimum threshold, 913 00:31:25,794 --> 00:31:27,606 maybe because of incidence of disease in 914 00:31:27,628 --> 00:31:29,926 these cohorts or something? Like do you 915 00:31:30,028 --> 00:31:31,914 not tend to look at anything less than 916 00:31:31,952 --> 00:31:33,850 10,000 samples or anything 917 00:31:33,920 --> 00:31:35,994 less than 50,000? Or do you look at 918 00:31:36,032 --> 00:31:38,970 each cohort on its own merit, based upon 919 00:31:38,970 --> 00:31:40,566 is there longitudinal 920 00:31:40,598 --> 00:31:42,142 data? How are the data 921 00:31:42,196 --> 00:31:44,894 collected? Can you share a few 922 00:31:44,932 --> 00:31:46,734 criteria that you think are important for 923 00:31:46,772 --> 00:31:48,906 selecting cohorts? That's 924 00:31:48,938 --> 00:31:50,746 a good question, and actually, we've 925 00:31:50,778 --> 00:31:52,814 started to think about this more 926 00:31:52,852 --> 00:31:54,686 objectively. Can we put together a list of 927 00:31:54,708 --> 00:31:56,474 criteria for biobank 928 00:31:56,522 --> 00:31:58,686 curation? Because now that the UK 929 00:31:58,718 --> 00:32:00,446 Biobank used to be the only game in town, 930 00:32:00,478 --> 00:32:02,498 but it's still probably, in my opinion, the 931 00:32:02,504 --> 00:32:03,966 best, but there are a lot of excellent 932 00:32:03,998 --> 00:32:05,686 cohorts coming out as well. 933 00:32:05,868 --> 00:32:07,798 The way that I think of it, and this is a 934 00:32:07,804 --> 00:32:09,734 little bit coarse and 935 00:32:09,772 --> 00:32:11,640 maybe crude, but 936 00:32:11,710 --> 00:32:13,894 the larger your 937 00:32:13,932 --> 00:32:15,862 sample size, the 938 00:32:15,996 --> 00:32:17,938 less detailed your phenotyping 939 00:32:17,954 --> 00:32:19,878 and your clinical information, and then the 940 00:32:19,884 --> 00:32:21,578 smaller the sample size, the more 941 00:32:21,664 --> 00:32:23,978 disease specific clinical phenotyping you 942 00:32:23,984 --> 00:32:25,898 can get. So I would say you could 943 00:32:25,904 --> 00:32:27,446 go all the way up to some of these medical 944 00:32:27,478 --> 00:32:29,866 records databases from Optum or 945 00:32:29,888 --> 00:32:31,466 IBM, and they've got hundreds of 946 00:32:31,488 --> 00:32:33,662 millions of people. And you can do some cool 947 00:32:33,716 --> 00:32:35,246 things with regard to 948 00:32:35,348 --> 00:32:37,566 comorbidity mapping in those databases, but 949 00:32:37,588 --> 00:32:39,422 you can't link to a specific 950 00:32:39,556 --> 00:32:41,614 clinical scale for depression or 951 00:32:41,652 --> 00:32:43,598 Alzheimer's disease and they're not going to 952 00:32:43,604 --> 00:32:45,310 have neuroimaging or 953 00:32:45,380 --> 00:32:47,678 proteomics, et cetera. And then on the 954 00:32:47,684 --> 00:32:49,298 other end of the spectrum, you have some of 955 00:32:49,304 --> 00:32:51,266 the cohorts that, like I mentioned at the 956 00:32:51,288 --> 00:32:53,426 start, the Swedish Neurology cohort that I 957 00:32:53,448 --> 00:32:55,782 was applying Olink to a few years ago 958 00:32:55,916 --> 00:32:57,826 that's got CDR [Clinical Dementia Rating] summary 959 00:32:57,858 --> 00:32:59,826 boxes and mini mental state examination 960 00:32:59,858 --> 00:33:01,794 and all of these very disease 961 00:33:01,842 --> 00:33:03,942 specific measurements that really 962 00:33:03,996 --> 00:33:05,590 help us drive in 963 00:33:05,660 --> 00:33:06,998 on specific 964 00:33:07,164 --> 00:33:09,494 hypotheses that are relevant to disease and 965 00:33:09,532 --> 00:33:11,546 sometimes almost use those studies like 966 00:33:11,568 --> 00:33:13,450 natural history cohorts or like control 967 00:33:13,520 --> 00:33:15,962 wings to clinical trials. And then you have 968 00:33:16,016 --> 00:33:18,166 sort of the, I won't say the "Goldilocks" 969 00:33:18,198 --> 00:33:20,426 biobank, but the goldilocks sort of 970 00:33:20,448 --> 00:33:22,214 approach, but I can't think of a better 971 00:33:22,272 --> 00:33:24,378 term. And that would be where the Biobanks 972 00:33:24,394 --> 00:33:26,900 fit in, I think UKB 973 00:33:26,900 --> 00:33:28,426 it doesn't capture everything, it doesn't 974 00:33:28,458 --> 00:33:30,798 have many mental state examination or 975 00:33:30,964 --> 00:33:32,834 CDR summary boxes. But it does 976 00:33:32,872 --> 00:33:34,622 have fluid intelligence 977 00:33:34,686 --> 00:33:36,866 tests, it has trail making. It has a lot of 978 00:33:36,888 --> 00:33:38,880 different cognitive and 979 00:33:38,880 --> 00:33:40,580 functional tasks, 980 00:33:40,580 --> 00:33:42,974 paired with deep 981 00:33:43,022 --> 00:33:45,686 genetic data. Now, proteomic data, 982 00:33:45,788 --> 00:33:47,906 actigraphy imaging, et cetera, 983 00:33:47,938 --> 00:33:49,730 et cetera. 984 00:33:49,730 --> 00:33:51,894 So, I think that finding that sort 985 00:33:51,932 --> 00:33:53,986 of goldilocks 986 00:33:54,018 --> 00:33:56,662 approach where we can get the power of large 987 00:33:56,716 --> 00:33:58,538 scale, but also get some of the 988 00:33:58,624 --> 00:34:00,714 denser clinical phenotyping, is 989 00:34:00,832 --> 00:34:02,646 usually how we try to go about it when we're 990 00:34:02,678 --> 00:34:04,742 selecting our cohorts. 991 00:34:04,886 --> 00:34:06,874 That's amazing. Wow. That was really 992 00:34:06,912 --> 00:34:08,880 a rich answer. Thanks. 993 00:34:10,670 --> 00:34:12,782 Well, Chris, it was great having you here 994 00:34:12,836 --> 00:34:14,746 on the podcast this afternoon. 995 00:34:14,910 --> 00:34:16,622 Thank you, really, so much 996 00:34:16,676 --> 00:34:18,718 for being so generous with your time and 997 00:34:18,724 --> 00:34:20,782 your thoughts today. We really look forward 998 00:34:20,836 --> 00:34:22,890 to seeing some results. 999 00:34:22,890 --> 00:34:24,862 And indeed, like you mentioned, 1000 00:34:24,916 --> 00:34:26,940 the creativity of scientists. 1001 00:34:26,940 --> 00:34:28,706 I'm very happy to be here. 1002 00:34:28,808 --> 00:34:30,546 Before I go, I do want to give a shout out 1003 00:34:30,568 --> 00:34:32,834 to Evan Mills again for helping get this 1004 00:34:32,872 --> 00:34:34,910 over the line and 1005 00:34:34,910 --> 00:34:36,962 to Klev Diamanti and 1006 00:34:37,016 --> 00:34:39,974 Philippa Pettingell, who did so 1007 00:34:40,012 --> 00:34:42,390 much technical and 1008 00:34:42,460 --> 00:34:44,806 just all sorts of technical support and 1009 00:34:44,828 --> 00:34:46,514 scientific support for the UKB 1010 00:34:46,562 --> 00:34:48,866 project. A disclaimer: 1011 00:34:48,978 --> 00:34:50,922 this was Chris's shout out to 1012 00:34:50,976 --> 00:34:52,810 three Olink employees in 1013 00:34:52,850 --> 00:34:54,986 recognition for all 1014 00:34:55,008 --> 00:34:57,146 their effort on this. But I'll also say that 1015 00:34:57,168 --> 00:34:59,834 I think Klev and Philippa have both said 1016 00:34:59,872 --> 00:35:01,886 how much they appreciated, how 1017 00:35:01,908 --> 00:35:03,854 much they learned from the 1018 00:35:03,892 --> 00:35:05,822 genetics perspective, from 1019 00:35:05,876 --> 00:35:07,822 so many of these thought leaders that are 1020 00:35:07,876 --> 00:35:09,854 the scientists in pharma who are 1021 00:35:09,892 --> 00:35:11,678 driving the experimental design and the 1022 00:35:11,684 --> 00:35:13,786 vision and gained 1023 00:35:13,818 --> 00:35:15,406 approval to use the UK Biobank 1024 00:35:15,438 --> 00:35:17,746 data. I think this idea of looking at 1025 00:35:17,768 --> 00:35:19,730 pharma as a funding body, 1026 00:35:19,880 --> 00:35:21,922 you shared that with me before, 1027 00:35:21,976 --> 00:35:23,918 Chris. These are heavy 1028 00:35:23,934 --> 00:35:25,294 hitting scientists that 1029 00:35:25,432 --> 00:35:27,174 have an 1030 00:35:27,292 --> 00:35:29,942 incredible track record of being able 1031 00:35:29,996 --> 00:35:31,542 to drive 1032 00:35:31,676 --> 00:35:33,778 such rich discoveries. 1033 00:35:33,874 --> 00:35:35,606 So it's such a 1034 00:35:35,628 --> 00:35:37,830 privilege to be around 1035 00:35:37,900 --> 00:35:39,798 you all and see this paper 1036 00:35:39,884 --> 00:35:41,890 coming out 1037 00:35:41,890 --> 00:35:43,530 from these data that 1038 00:35:43,680 --> 00:35:45,914 you've all been a part of. So I look forward 1039 00:35:45,952 --> 00:35:47,882 to that publication, too, in case 1040 00:35:47,936 --> 00:35:49,418 you can plug for it. I don't know if you 1041 00:35:49,424 --> 00:35:50,846 know any timing around the 1042 00:35:50,948 --> 00:35:52,718 UK-PPP paper. 1043 00:35:52,804 --> 00:35:54,606 First paper. I 1044 00:35:54,628 --> 00:35:56,558 resubmitted the revised version over the 1045 00:35:56,564 --> 00:35:58,980 weekend. 1046 00:35:58,980 --> 00:36:00,894 Someone else 1047 00:36:00,932 --> 00:36:02,890 was working over the weekend then. 1048 00:36:02,890 --> 00:36:04,894 The response to the reviewers 1049 00:36:04,942 --> 00:36:06,898 was 29 pages long. That can either be a 1050 00:36:06,904 --> 00:36:08,870 good thing or a bad thing 1051 00:36:10,870 --> 00:36:12,670 A lot of novel methods, I think. 1052 00:36:12,670 --> 00:36:14,526 Well, that's 1053 00:36:14,558 --> 00:36:16,358 exciting. That's exciting. You heard it 1054 00:36:16,364 --> 00:36:18,774 first here. Yes. Plenty to look forward 1055 00:36:18,812 --> 00:36:20,934 to. Thanks again very much, 1056 00:36:20,972 --> 00:36:22,598 Chris, it was great. Thank you. 1057 00:36:22,684 --> 00:36:24,860 Yeah, good to be here. 1058 00:36:28,860 --> 00:36:30,838 Thank you for listening to the 1059 00:36:30,844 --> 00:36:32,706 Proteomics in Proximity podcast 1060 00:36:32,818 --> 00:36:34,898 brought to you by Olink Proteomics. 1061 00:36:34,994 --> 00:36:36,754 To contact the hosts or for further 1062 00:36:36,802 --> 00:36:38,438 information, simply email 1063 00:36:38,524 --> 00:36:40,970 info@olink.com.