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 Yazuki, 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 --> 00:00:24,998 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 --> 00:00:29,894 your hosts, Dale, Cindy, and 12 00:00:29,932 --> 00:00:31,666 Sarantis. Hello, everybody. I'm 13 00:00:31,698 --> 00:00:33,874 Sarantis. I'm together today with Dale 14 00:00:33,922 --> 00:00:35,910 and Cindy for another episode 15 00:00:35,910 --> 00:00:37,746 of our 16 00:00:37,788 --> 00:00:39,546 great podcast, Proteomics in 17 00:00:39,568 --> 00:00:41,850 Proximity. We are all very happy 18 00:00:41,920 --> 00:00:43,734 to have like a guest, Professor 19 00:00:43,782 --> 00:00:45,846 Johann Schwenk 20 00:00:45,878 --> 00:00:47,834 who holds a 21 00:00:47,872 --> 00:00:49,878 position at the University of 22 00:00:49,984 --> 00:00:51,330 KTH University [Royal Institute of Technology in Sweden]. 23 00:00:51,330 --> 00:00:53,880 And today he's 24 00:00:53,880 --> 00:00:55,902 a protein expert and 25 00:00:55,956 --> 00:00:57,966 a professor in 26 00:00:57,988 --> 00:00:59,866 Translational Proteomics. And today we'll 27 00:00:59,898 --> 00:01:01,870 discuss a little bit about his 28 00:01:01,940 --> 00:01:03,634 new research, about his 29 00:01:03,672 --> 00:01:05,506 research interest and how proteins can 30 00:01:05,528 --> 00:01:07,950 enable multiomics approaches. Actually, 31 00:01:07,950 --> 00:01:09,600 Jochen, 32 00:01:09,600 --> 00:01:11,454 thank you very much for joining 33 00:01:11,582 --> 00:01:13,750 today. And I would like to 34 00:01:13,750 --> 00:01:15,986 start the discussion asking you: What 35 00:01:16,008 --> 00:01:18,770 does translational proteomics mean to you? 36 00:01:20,570 --> 00:01:22,820 Yeah, 37 00:01:22,820 --> 00:01:24,934 I think when we started thinking 38 00:01:24,972 --> 00:01:26,086 about the title for a 39 00:01:26,108 --> 00:01:28,646 professorship, translation was really a hot 40 00:01:28,668 --> 00:01:30,760 topic at the time: to bring 41 00:01:30,760 --> 00:01:32,626 something you've been doing in the lab 42 00:01:32,668 --> 00:01:34,586 into a clinical context. But I think 43 00:01:34,608 --> 00:01:36,826 it turns out to be much more than this. This is 44 00:01:36,848 --> 00:01:38,838 actually to explain also what we're 45 00:01:38,854 --> 00:01:40,574 doing in the lab to others, so that 46 00:01:40,692 --> 00:01:42,782 the community can engage into 47 00:01:42,836 --> 00:01:44,686 our research and we can even find a 48 00:01:44,708 --> 00:01:46,586 broader utility. So it's 49 00:01:46,618 --> 00:01:48,890 still the idea of 50 00:01:48,890 --> 00:01:50,874 connecting the lab environment 51 00:01:50,922 --> 00:01:52,794 with clinical and population 52 00:01:52,842 --> 00:01:54,962 health. So I think 53 00:01:55,016 --> 00:01:57,860 hopefully one day we'll contribute to that 54 00:01:57,860 --> 00:01:59,842 activity. That's great. 55 00:01:59,896 --> 00:02:01,886 And I saw also that you study biochemistry 56 00:02:01,918 --> 00:02:03,506 in Tübingen, University of Tübingen. That's quite 57 00:02:03,528 --> 00:02:05,986 famous from the biochemistry industry worldwide. Do 58 00:02:06,008 --> 00:02:07,366 you have any story that you would like to 59 00:02:07,388 --> 00:02:09,930 record for your first paper? For example, 60 00:02:09,930 --> 00:02:11,686 being in Tübingen, and that would be great 61 00:02:11,708 --> 00:02:13,940 to hear. 62 00:02:13,940 --> 00:02:15,920 Yeah, 63 00:02:15,920 --> 00:02:17,878 Tübingen of course gave me a 64 00:02:17,884 --> 00:02:19,734 fantastic time. We were very 65 00:02:19,772 --> 00:02:21,946 small number of students per semester. I had 66 00:02:21,968 --> 00:02:23,754 very close connection to the 67 00:02:23,792 --> 00:02:25,946 professors. I got a chance to go to 68 00:02:25,968 --> 00:02:27,700 a Lindau Nobel Laureate 69 00:02:27,700 --> 00:02:29,994 meeting there. And it was really, 70 00:02:30,032 --> 00:02:32,780 I think, an inspirational time that 71 00:02:32,780 --> 00:02:34,886 sort of created 72 00:02:34,918 --> 00:02:36,766 a lot of curiosity about science. And 73 00:02:36,788 --> 00:02:38,606 then after that I moved a little bit more 74 00:02:38,628 --> 00:02:40,686 into technology. So in the early two 75 00:02:40,708 --> 00:02:42,686 2000s, when I did my Masters and PhD, I 76 00:02:42,708 --> 00:02:44,174 worked with Luminex-based 77 00:02:44,212 --> 00:02:46,642 assays, which at that time 78 00:02:46,696 --> 00:02:48,850 was really new. 79 00:02:48,850 --> 00:02:50,802 Then sort of that took me then 80 00:02:50,856 --> 00:02:52,626 to join the Protein Atlas in 81 00:02:52,648 --> 00:02:54,866 2005 as a postdoc. And somehow I 82 00:02:54,888 --> 00:02:56,920 got stuck with this fantastic project. 83 00:02:56,950 --> 00:02:58,866 I'm 84 00:02:58,898 --> 00:03:00,966 still around and 85 00:03:01,068 --> 00:03:02,694 learn every day something new about 86 00:03:02,732 --> 00:03:04,774 proteins. I mean, just being 87 00:03:04,812 --> 00:03:06,898 there and to work with Mathias Uhlén 88 00:03:06,914 --> 00:03:08,518 and all the colleagues has been truly 89 00:03:08,604 --> 00:03:10,326 inspirational. So you've been at 90 00:03:10,348 --> 00:03:12,986 SciLife Labs since 2005, and that was when 91 00:03:13,008 --> 00:03:15,760 it began at KTH, is that correct? 92 00:03:15,760 --> 00:03:17,706 Yes. SciLife lab was 93 00:03:17,728 --> 00:03:19,850 inaugurated in 2010. So actually, 94 00:03:19,920 --> 00:03:21,846 it was my and three other groups 95 00:03:21,878 --> 00:03:23,902 that moved into the building under 96 00:03:23,956 --> 00:03:25,838 construction, I think it was in 97 00:03:25,924 --> 00:03:27,294 October 2010. 98 00:03:27,412 --> 00:03:29,966 So I consider myself very much of an 99 00:03:29,988 --> 00:03:31,806 oldie when I think about my time at 100 00:03:31,828 --> 00:03:33,566 SciLife Lab. I've seen it 101 00:03:33,668 --> 00:03:35,698 change, grow, and now, I think, 102 00:03:35,784 --> 00:03:37,966 become a very prominent research institute 103 00:03:37,998 --> 00:03:39,778 in Europe. So it's, I 104 00:03:39,784 --> 00:03:41,746 think, fantastic and very 105 00:03:41,768 --> 00:03:43,954 much also gave me opportunities to learn 106 00:03:43,992 --> 00:03:45,826 about other technologies and to 107 00:03:45,928 --> 00:03:47,974 learn how information about 108 00:03:48,012 --> 00:03:50,806 proteins can be useful. And I 109 00:03:50,828 --> 00:03:52,854 have many stories to tell, but 110 00:03:52,892 --> 00:03:54,518 one of which is, for instance, I have a 111 00:03:54,524 --> 00:03:56,950 little bit of a side activity project around 112 00:03:57,020 --> 00:03:59,846 GPCRs, and that, for instance, I think 113 00:03:59,868 --> 00:04:01,434 wouldn't have been possible if I would just 114 00:04:01,472 --> 00:04:03,870 be sitting somewhere in a lab and 115 00:04:03,870 --> 00:04:05,562 not be exposed to all these different 116 00:04:05,616 --> 00:04:07,274 activities. Sure. And for those not 117 00:04:07,312 --> 00:04:09,702 familiar, GPCRs 118 00:04:09,702 --> 00:04:11,610 or G- 119 00:04:11,610 --> 00:04:13,370 coupled protein receptors. 120 00:04:13,450 --> 00:04:15,860 GCPRs, is that correct? 121 00:04:15,860 --> 00:04:17,946 G-protein coupled 122 00:04:17,978 --> 00:04:19,966 receptors. G-protein coupled receptors. I 123 00:04:19,988 --> 00:04:21,880 got to get my acronym straight. 124 00:04:21,880 --> 00:04:23,454 It's a really important drug 125 00:04:23,502 --> 00:04:25,886 target, right? Membrane 126 00:04:25,918 --> 00:04:27,506 proteins. Yeah, membrane proteins that are 127 00:04:27,528 --> 00:04:29,790 important drug targets. Correct? 128 00:04:29,790 --> 00:04:31,714 Exactly. Yeah. 129 00:04:31,752 --> 00:04:33,326 And they're at the SciLife 130 00:04:33,358 --> 00:04:34,978 Laboratories. Well, you said that you were 131 00:04:34,984 --> 00:04:36,994 involved in the Human Protein Atlas way back 132 00:04:37,032 --> 00:04:39,794 in 2005, so therefore 133 00:04:39,922 --> 00:04:41,814 the genome had just been 134 00:04:41,852 --> 00:04:42,854 finished in 2002. 135 00:04:42,892 --> 00:04:44,970 2003. 136 00:04:44,970 --> 00:04:46,754 Those must have been pretty exciting 137 00:04:46,802 --> 00:04:48,726 times because there was a big pivot 138 00:04:48,758 --> 00:04:50,506 and interest and focus on the 139 00:04:50,528 --> 00:04:52,910 proteome, is that correct? 140 00:04:52,910 --> 00:04:54,906 Yeah, of course. And then at 141 00:04:54,928 --> 00:04:56,790 that time, Mathias 142 00:04:56,790 --> 00:04:58,746 and us were producing all these 143 00:04:58,768 --> 00:05:00,554 antibodies was fairly unique, and people 144 00:05:00,592 --> 00:05:02,366 were not really sure whether that would add 145 00:05:02,388 --> 00:05:04,974 any value to the field 146 00:05:05,012 --> 00:05:07,646 that's dominated by mass spectrometry. But I 147 00:05:07,668 --> 00:05:09,810 think now we've shown, and 148 00:05:09,810 --> 00:05:11,774 the way that we've brought in new 149 00:05:11,812 --> 00:05:13,874 data, trying to understand the data that we 150 00:05:13,912 --> 00:05:15,726 generate, and then sort of give feedback 151 00:05:15,838 --> 00:05:17,762 to other data types with 152 00:05:17,816 --> 00:05:18,798 localization of 153 00:05:18,974 --> 00:05:20,760 subcellular compartments 154 00:05:20,760 --> 00:05:22,802 of proteins that 155 00:05:22,856 --> 00:05:24,974 I think are really 156 00:05:25,112 --> 00:05:27,730 super valuable and help us to disentangle 157 00:05:27,810 --> 00:05:29,814 the complex biology that we 158 00:05:29,852 --> 00:05:31,900 live in. And 159 00:05:31,900 --> 00:05:33,798 my real interest is 160 00:05:33,884 --> 00:05:35,766 proteins in the circulation. So that's even 161 00:05:35,788 --> 00:05:37,698 more complicated because 162 00:05:37,884 --> 00:05:39,322 you're under sort of the 163 00:05:39,456 --> 00:05:41,894 constant exchange of molecules 164 00:05:41,942 --> 00:05:43,962 in all parts of the body. 165 00:05:44,016 --> 00:05:46,426 So it's not as organized as looking at 166 00:05:46,448 --> 00:05:48,618 subcellular localizations, but it's still 167 00:05:48,704 --> 00:05:50,980 fascinating. And I guess 168 00:05:50,980 --> 00:05:52,974 that's something I really 169 00:05:53,012 --> 00:05:55,886 sort of fell in love with, and I 170 00:05:55,908 --> 00:05:57,790 really enjoy doing. 171 00:05:57,790 --> 00:05:59,890 Great. 172 00:05:59,890 --> 00:06:01,934 How would you see Olink because 173 00:06:01,972 --> 00:06:03,758 you are a biochemist? I 174 00:06:03,764 --> 00:06:05,690 mean, you are a mass spec 175 00:06:05,690 --> 00:06:07,694 expert. How will you see 176 00:06:07,732 --> 00:06:09,458 Olink fitting on this pipeline of 177 00:06:09,464 --> 00:06:11,106 mass spectrometry? How do you see mass spec and Olink 178 00:06:11,128 --> 00:06:12,530 working together from your experience? 179 00:06:12,600 --> 00:06:13,986 Because you have a big experience in this 180 00:06:14,008 --> 00:06:16,790 field. 181 00:06:16,790 --> 00:06:18,970 I'm very fortunate to get to know Ulf Landegren 182 00:06:18,990 --> 00:06:20,854 many, many years back, and we've been 183 00:06:20,892 --> 00:06:22,994 sort of seeing each other on a regular 184 00:06:23,042 --> 00:06:25,846 basis, partly because Uppsala [Olink] and Sweden [SciLife Labs] are 185 00:06:25,868 --> 00:06:27,782 very close and sort of been 186 00:06:27,836 --> 00:06:29,514 doing things in parallel. And of course, 187 00:06:29,552 --> 00:06:31,834 fantastic to see the journey that with the 188 00:06:31,872 --> 00:06:33,894 proximity extension, proximity ligation, 189 00:06:33,942 --> 00:06:35,550 and all these different 190 00:06:35,550 --> 00:06:37,754 versions of this concept have 191 00:06:37,792 --> 00:06:39,930 now, I think, 192 00:06:40,080 --> 00:06:42,838 been the driver for using antibodies as 193 00:06:42,864 --> 00:06:44,846 molecular tools. I mean, there was 194 00:06:44,868 --> 00:06:46,446 just a paper in Nature Methods, I think, 195 00:06:46,468 --> 00:06:48,490 just the other day. Again antibodies 196 00:06:48,570 --> 00:06:50,734 conjugated with oligonucleotides. I think 197 00:06:50,772 --> 00:06:52,830 that's giving people much more 198 00:06:52,932 --> 00:06:54,754 bigger field to play and use these 199 00:06:54,792 --> 00:06:56,674 reagents. So 200 00:06:56,712 --> 00:06:58,994 of course I've seen how it 201 00:06:59,032 --> 00:07:01,618 started and we've been 202 00:07:01,704 --> 00:07:03,682 very late to the game. My lab, 203 00:07:03,736 --> 00:07:05,774 or the unit that I'm heading 204 00:07:05,822 --> 00:07:07,846 at SciLife Lab, started to 205 00:07:07,868 --> 00:07:09,366 introduce Olink in 2017. 206 00:07:09,388 --> 00:07:11,734 And since then, we've been super 207 00:07:11,772 --> 00:07:13,430 happy to have the system in-house 208 00:07:13,500 --> 00:07:15,810 and do this 209 00:07:15,810 --> 00:07:17,894 for others, users that 210 00:07:17,932 --> 00:07:19,958 come to SciLife Lab to just want to 211 00:07:19,964 --> 00:07:21,754 have data, but also for our 212 00:07:21,792 --> 00:07:23,994 research. So I think, again, any 213 00:07:24,032 --> 00:07:26,698 data type adds a value to what we do. And I 214 00:07:26,704 --> 00:07:28,510 think Olink has truly 215 00:07:28,510 --> 00:07:30,986 enabled us to do many things we weren't 216 00:07:31,018 --> 00:07:33,170 able to before. So, 217 00:07:33,170 --> 00:07:35,866 fantastic. I'm dying 218 00:07:35,898 --> 00:07:37,890 to ask 219 00:07:37,890 --> 00:07:39,550 in front of you, 220 00:07:39,700 --> 00:07:41,854 based on what you see, the 221 00:07:41,892 --> 00:07:43,780 opportunity in front of you 222 00:07:43,780 --> 00:07:45,490 with this technology, whatever 223 00:07:45,560 --> 00:07:47,806 technologies, right. What is it that you're 224 00:07:47,838 --> 00:07:49,666 most excited about for the 225 00:07:49,688 --> 00:07:51,374 future? Are there aspects 226 00:07:51,422 --> 00:07:53,906 of 227 00:07:53,928 --> 00:07:55,846 work that you've been doing or a 228 00:07:55,868 --> 00:07:57,830 direction that you're going in, that 229 00:07:57,980 --> 00:07:59,474 you would like to share, that you're 230 00:07:59,522 --> 00:08:01,910 comfortable sharing: 231 00:08:01,910 --> 00:08:03,690 I just want to know 232 00:08:03,690 --> 00:08:05,462 what's the part that makes 233 00:08:05,516 --> 00:08:07,670 you go into 234 00:08:07,740 --> 00:08:09,958 flow? What do 235 00:08:09,964 --> 00:08:11,890 you want to do next? 236 00:08:13,690 --> 00:08:15,946 Yeah, I started 237 00:08:16,048 --> 00:08:18,350 doing a lot of assay development myself 238 00:08:18,350 --> 00:08:20,634 when I worked with Luminix 20 years 239 00:08:20,672 --> 00:08:22,526 back. That sounds a little bit silly when I 240 00:08:22,548 --> 00:08:24,960 say this, but it's the truth. 241 00:08:24,960 --> 00:08:26,878 Of course, that sort of 242 00:08:26,884 --> 00:08:28,782 has always been something to try, something 243 00:08:28,836 --> 00:08:30,266 new to maybe try, something that's 244 00:08:30,298 --> 00:08:32,810 difficult, maybe not immediately 245 00:08:32,890 --> 00:08:34,926 rewarding, but in the 246 00:08:34,948 --> 00:08:36,458 long term, something that could 247 00:08:36,484 --> 00:08:38,946 be very fruitful or something that 248 00:08:38,968 --> 00:08:40,626 makes you proud as the researchers that you 249 00:08:40,648 --> 00:08:42,386 say [to yourself]: "Okay, this is something I believed in, 250 00:08:42,408 --> 00:08:44,866 and I see it's happening." So the 251 00:08:44,888 --> 00:08:46,946 next sort of moment for me, 252 00:08:46,968 --> 00:08:48,934 when I had this type of thinking was 253 00:08:48,972 --> 00:08:50,854 when COVID started and when 254 00:08:50,892 --> 00:08:52,850 lots of people went into serology 255 00:08:52,930 --> 00:08:54,918 testing or protein testing in 256 00:08:54,924 --> 00:08:56,966 the classical way, when me 257 00:08:56,988 --> 00:08:58,742 and my colleagues at KTH, we said, 258 00:08:58,876 --> 00:09:00,806 "Let's try something different and use dried 259 00:09:00,838 --> 00:09:02,794 blood spots. Let's not ask 260 00:09:02,832 --> 00:09:04,530 people to come to the clinic, let's send 261 00:09:04,530 --> 00:09:06,780 the devices back to them, 262 00:09:06,930 --> 00:09:08,934 to their home, so they can collect 263 00:09:08,982 --> 00:09:10,822 bloods in their kitchen 264 00:09:10,886 --> 00:09:12,954 and sofa, wherever, and send 265 00:09:12,992 --> 00:09:14,910 them the samples back to us, to the lab, 266 00:09:14,980 --> 00:09:16,820 where we can do the research." 267 00:09:16,820 --> 00:09:18,574 So that, I think, really 268 00:09:18,612 --> 00:09:20,990 inspired a lot of new ways of 269 00:09:21,060 --> 00:09:23,326 doing this. When you think about cutting 270 00:09:23,358 --> 00:09:25,966 costs, simplifying workflows, freeing 271 00:09:25,998 --> 00:09:27,854 the time of people in the clinic, 272 00:09:27,982 --> 00:09:29,570 but also to think about 273 00:09:29,720 --> 00:09:31,746 doing health monitoring. I 274 00:09:31,768 --> 00:09:33,590 think people 275 00:09:33,590 --> 00:09:35,842 always often ask me, what do I think is 276 00:09:35,896 --> 00:09:37,762 proteomics best used 277 00:09:37,816 --> 00:09:39,986 for? And I think it's monitoring. 278 00:09:40,018 --> 00:09:42,262 It's like looking at who you are and looking 279 00:09:42,316 --> 00:09:44,966 how you change. And I think this 280 00:09:44,988 --> 00:09:46,678 combination, I think, now is sort 281 00:09:46,684 --> 00:09:48,918 of shaping towards something I really 282 00:09:49,004 --> 00:09:51,446 am passionate about. And dried blood spots 283 00:09:51,478 --> 00:09:53,790 is a fantastic tool. 284 00:09:53,790 --> 00:09:55,866 It's more challenging than doing it 285 00:09:55,888 --> 00:09:57,706 the classical way. There's so much more to 286 00:09:57,728 --> 00:09:59,594 learn, and then maybe even go 287 00:09:59,632 --> 00:10:01,962 further. When 288 00:10:02,016 --> 00:10:04,126 Sarantis and I talked a couple of 289 00:10:04,148 --> 00:10:06,666 weeks back, also to look at even smaller 290 00:10:06,698 --> 00:10:08,574 sample volumes, looking at other body 291 00:10:08,612 --> 00:10:10,234 fluids, such as interstitial 292 00:10:10,282 --> 00:10:12,974 fluid, that could even tell us something 293 00:10:13,012 --> 00:10:15,950 extra that blood is not able to tell us. 294 00:10:15,950 --> 00:10:17,938 We'd need a 295 00:10:17,944 --> 00:10:19,986 baseline to understand the reference of 296 00:10:20,008 --> 00:10:22,594 dried blood spots to be making that 297 00:10:22,632 --> 00:10:24,690 comparison, right? Yeah. And 298 00:10:24,760 --> 00:10:26,966 so enabling in areas where we 299 00:10:26,988 --> 00:10:28,770 can't get a blood draw, 300 00:10:28,770 --> 00:10:30,918 a phlebotomist out to do a 301 00:10:30,924 --> 00:10:32,790 blood draw. 302 00:10:32,790 --> 00:10:34,774 I think that's going to be really 303 00:10:34,812 --> 00:10:36,810 important. Dale? 304 00:10:36,810 --> 00:10:38,994 So if you can give us some background 305 00:10:39,042 --> 00:10:41,538 on dried blood spots, and I'd appreciate it, 306 00:10:41,564 --> 00:10:43,658 because my only familiarity with it was when 307 00:10:43,664 --> 00:10:45,978 I had my first child, and they did a 308 00:10:45,984 --> 00:10:47,914 heel prick, and then they went ahead 309 00:10:47,952 --> 00:10:49,914 and used the blood from 310 00:10:49,952 --> 00:10:51,878 that little lancet onto a 311 00:10:51,904 --> 00:10:53,742 particular card. Is there something 312 00:10:53,796 --> 00:10:55,774 special about the material they 313 00:10:55,812 --> 00:10:57,800 use for a dried blood spot? 314 00:10:57,800 --> 00:10:59,998 And what are the challenges as far 315 00:11:00,004 --> 00:11:02,770 as working with proteins in that context? 316 00:11:02,770 --> 00:11:04,960 I think 317 00:11:04,960 --> 00:11:06,946 let's say if you think 318 00:11:06,968 --> 00:11:08,814 from an analytical 319 00:11:08,942 --> 00:11:10,254 perspective and a precision 320 00:11:10,302 --> 00:11:12,914 perspective, the dry blood spots your 321 00:11:12,952 --> 00:11:14,950 kids 322 00:11:14,950 --> 00:11:16,978 donate, they are just put in a 323 00:11:16,984 --> 00:11:18,854 filter paper to do a plus/minus test. 324 00:11:18,892 --> 00:11:20,866 So it's really a binary answer you're 325 00:11:20,898 --> 00:11:22,918 after. But if you really want to look 326 00:11:22,924 --> 00:11:24,742 at subtle changes in the human 327 00:11:24,876 --> 00:11:26,806 phenotype, I think then you need 328 00:11:26,828 --> 00:11:28,726 to ensure that the precision of the 329 00:11:28,748 --> 00:11:30,726 material you use in your system is 330 00:11:30,748 --> 00:11:32,590 there. 331 00:11:32,590 --> 00:11:34,266 Normally when you have a dry blood spot, you 332 00:11:34,288 --> 00:11:36,234 get sort of a donut distribution of the red 333 00:11:36,272 --> 00:11:38,458 blood cells, so it really matters where you 334 00:11:38,464 --> 00:11:40,954 do the punch. So you want to avoid these 335 00:11:40,992 --> 00:11:42,718 type of things, especially if you want to do 336 00:11:42,724 --> 00:11:44,334 it at scale, if you want to do it 337 00:11:44,372 --> 00:11:46,920 consecutively. So I 338 00:11:47,012 --> 00:11:49,534 started to work with a local company that 339 00:11:49,572 --> 00:11:51,790 was founded by one of my colleagues at 340 00:11:51,860 --> 00:11:53,998 KTH. And they use a 341 00:11:54,004 --> 00:11:56,894 microfuidic system to exactly collect 342 00:11:56,942 --> 00:11:58,878 ten microliters. So just knowing 343 00:11:58,894 --> 00:12:00,754 that what you put into your system is ten 344 00:12:00,792 --> 00:12:02,546 microliters, of course, 345 00:12:02,648 --> 00:12:04,386 then there are different levels of 346 00:12:04,408 --> 00:12:05,954 hematocrits, there are different other 347 00:12:05,992 --> 00:12:07,990 things that you need to consider, but you 348 00:12:08,060 --> 00:12:10,326 at least eliminate some of the concerns that 349 00:12:10,348 --> 00:12:12,966 you have. That, 350 00:12:12,988 --> 00:12:14,966 I think, is really the key. And of course, 351 00:12:14,988 --> 00:12:16,854 it's a simplicity of this 352 00:12:16,892 --> 00:12:18,886 procedure that you can assure it's 353 00:12:18,918 --> 00:12:20,954 easy for people to do. And 354 00:12:20,992 --> 00:12:22,986 they manage, even though they may not 355 00:12:23,088 --> 00:12:25,914 be trained. I failed also 356 00:12:25,952 --> 00:12:27,940 when I did it the first times. 357 00:12:27,940 --> 00:12:29,786 But if you 358 00:12:29,808 --> 00:12:31,786 get used to it, the quality is 359 00:12:31,808 --> 00:12:33,694 really excellent. And I think there are also 360 00:12:33,732 --> 00:12:35,822 studies showing that more and more are using 361 00:12:35,876 --> 00:12:37,860 other devices 362 00:12:37,890 --> 00:12:39,486 that are out there, I 363 00:12:39,508 --> 00:12:41,738 think now you have a new material 364 00:12:41,834 --> 00:12:43,962 which is sort of similar to plasma, 365 00:12:44,026 --> 00:12:46,626 but it has some bonus. And the question is, 366 00:12:46,648 --> 00:12:48,994 how do you manage that bonus? Is 367 00:12:49,032 --> 00:12:50,658 it something that is a challenge, it's a 368 00:12:50,664 --> 00:12:52,626 burden that makes it difficult for 369 00:12:52,648 --> 00:12:54,870 you 370 00:12:54,870 --> 00:12:56,642 to be analytically 371 00:12:56,706 --> 00:12:58,614 precise? Or does it open up 372 00:12:58,652 --> 00:13:00,742 opportunities that were not possible 373 00:13:00,876 --> 00:13:02,914 when you looked at the regular blood, 374 00:13:02,940 --> 00:13:04,870 plasma samples? Because of, 375 00:13:04,940 --> 00:13:06,326 let's say, the 376 00:13:06,348 --> 00:13:08,134 hematopoietic cells that are still there, 377 00:13:08,172 --> 00:13:09,734 they may leak out something that could be 378 00:13:09,772 --> 00:13:11,958 really exciting. So it's 379 00:13:11,974 --> 00:13:13,600 this balance between things that, I think ... 380 00:13:13,600 --> 00:13:15,274 When you have a discussion a little bit 381 00:13:15,312 --> 00:13:17,418 with other proteomics expert about 382 00:13:17,424 --> 00:13:19,562 dry blood spots, there's always a question 383 00:13:19,616 --> 00:13:21,418 about how you control and 384 00:13:21,424 --> 00:13:23,760 normalize. Because, I don't know, 385 00:13:23,760 --> 00:13:25,406 from what I have heard, 386 00:13:25,508 --> 00:13:27,822 actually, it's not easy to have always 387 00:13:27,876 --> 00:13:29,978 the same type of dry blood spots. I'm 388 00:13:29,994 --> 00:13:31,994 guessing that there's a lot of varieties, 389 00:13:32,042 --> 00:13:34,158 a lot of variation can be there. Would you 390 00:13:34,164 --> 00:13:36,306 have any idea how one can normalize this 391 00:13:36,328 --> 00:13:37,982 data in order to have like, longitudinal 392 00:13:38,046 --> 00:13:40,978 studies or studies for different cohorts? Do 393 00:13:40,984 --> 00:13:42,626 you have any idea on that? That would be 394 00:13:42,648 --> 00:13:44,860 great to hear, actually. 395 00:13:44,860 --> 00:13:46,994 Yeah. I mean, there are analytical 396 00:13:47,042 --> 00:13:48,854 concepts that you can think about to do 397 00:13:48,892 --> 00:13:50,902 precision. You, probably similar 398 00:13:50,956 --> 00:13:52,806 to other studies, try to 399 00:13:52,828 --> 00:13:54,578 find some housekeeping 400 00:13:54,754 --> 00:13:56,934 markers. And we found some, for 401 00:13:56,972 --> 00:13:58,980 instance, that are related to skin. 402 00:13:58,980 --> 00:14:00,940 The skin, when you do 403 00:14:00,940 --> 00:14:02,600 the landset, 404 00:14:02,600 --> 00:14:04,598 or punching through your upper 405 00:14:04,694 --> 00:14:06,826 layers of the skin, these proteins will 406 00:14:06,848 --> 00:14:08,860 probably always be there. 407 00:14:08,990 --> 00:14:10,874 So trying to figure out 408 00:14:10,992 --> 00:14:12,746 what are the markers that are constant. And 409 00:14:12,768 --> 00:14:14,814 then again, what we talked about 410 00:14:14,852 --> 00:14:16,846 before, if you have a phenotype that 411 00:14:16,868 --> 00:14:18,974 is changing, then you can sort of do this 412 00:14:19,012 --> 00:14:20,558 resampling. You can learn from the 413 00:14:20,564 --> 00:14:22,494 resampling what are the 414 00:14:22,532 --> 00:14:24,426 constant constituents and what are those 415 00:14:24,468 --> 00:14:26,946 that are variable, what are those that are 416 00:14:26,968 --> 00:14:28,710 unreliable. 417 00:14:28,710 --> 00:14:30,914 And again, the more data you have, the 418 00:14:30,952 --> 00:14:32,798 easier it is to make that exercise, 419 00:14:32,894 --> 00:14:34,882 because you can rank things 420 00:14:34,936 --> 00:14:36,990 in a much more refined way. 421 00:14:38,790 --> 00:14:40,610 That's very knowledgeable. You mean 422 00:14:40,680 --> 00:14:42,486 using housekeeping kind of 423 00:14:42,508 --> 00:14:44,326 housekeeping proteins in a way, right, to 424 00:14:44,348 --> 00:14:46,886 normalize? That's pretty much the idea 425 00:14:46,988 --> 00:14:48,970 around? Yeah, exactly. 426 00:14:49,040 --> 00:14:50,954 And then, of course, it's just also a matter 427 00:14:50,992 --> 00:14:52,954 of using different 428 00:14:52,980 --> 00:14:54,634 statistical models to do 429 00:14:54,752 --> 00:14:56,506 normalization and things like this. I 430 00:14:56,528 --> 00:14:58,906 mean, whenever you have 431 00:14:58,928 --> 00:15:00,926 a variable sample source, I 432 00:15:00,948 --> 00:15:02,718 guess we have that also in 433 00:15:02,724 --> 00:15:04,530 plasma, different 434 00:15:04,530 --> 00:15:06,714 degrees of hemolysis, different fat 435 00:15:06,762 --> 00:15:08,734 content, different 436 00:15:08,772 --> 00:15:10,974 hydration states, they can 437 00:15:11,012 --> 00:15:13,474 influence so many things. So I think just 438 00:15:13,512 --> 00:15:15,778 keeping being on your toes when you look at 439 00:15:15,784 --> 00:15:17,694 the data and not get carried 440 00:15:17,742 --> 00:15:19,900 away too quickly, 441 00:15:19,900 --> 00:15:21,840 I think 442 00:15:21,990 --> 00:15:23,998 something that's very helpful. 443 00:15:24,174 --> 00:15:26,534 Coming back again, and I'm sorry I 444 00:15:26,572 --> 00:15:28,726 monopolized the questions, coming back 445 00:15:28,748 --> 00:15:30,806 again to Cindy's 446 00:15:30,828 --> 00:15:32,486 question: Do you think there will be a 447 00:15:32,508 --> 00:15:34,502 new breakthrough? You'll be like going 448 00:15:34,636 --> 00:15:36,660 through new matrices, like 449 00:15:36,660 --> 00:15:38,566 interstitial fluid, for 450 00:15:38,588 --> 00:15:40,306 example, do you think that new materials 451 00:15:40,338 --> 00:15:42,722 will open new ways, new research areas, 452 00:15:42,786 --> 00:15:44,746 and learn quite a lot? What is your 453 00:15:44,768 --> 00:15:46,900 feeling about that? What is your vision? 454 00:15:48,270 --> 00:15:50,666 I 455 00:15:50,688 --> 00:15:52,874 think we should accept 456 00:15:52,874 --> 00:15:54,922 the concept that not all material 457 00:15:54,986 --> 00:15:56,782 will be informative for all 458 00:15:56,836 --> 00:15:58,926 studies we do. So I think if we 459 00:15:58,948 --> 00:16:00,894 find the niche, that they are 460 00:16:00,932 --> 00:16:02,814 informative. So again, this 461 00:16:02,852 --> 00:16:04,994 study we did on interstitial fluid, we 462 00:16:05,032 --> 00:16:07,054 also detected, for instance, antibodies 463 00:16:07,102 --> 00:16:09,246 against SARS-CoV-2 in interstitial 464 00:16:09,278 --> 00:16:11,954 fluid. And then of course, that 465 00:16:11,992 --> 00:16:13,666 is a proof-of-concept we did 466 00:16:13,768 --> 00:16:15,826 because we were curious and we had 467 00:16:15,848 --> 00:16:17,798 material, or we had plus/minus as the 468 00:16:17,804 --> 00:16:19,858 phenotype. But imagine you're 469 00:16:19,874 --> 00:16:21,878 treating someone with melanoma, with a 470 00:16:21,884 --> 00:16:23,922 biologics. How can you assure 471 00:16:23,986 --> 00:16:25,462 that the biologic actually 472 00:16:25,516 --> 00:16:27,830 reaches the area where it should act? 473 00:16:27,980 --> 00:16:29,974 I think these things could, 474 00:16:30,012 --> 00:16:32,790 of course, be much more informative than 475 00:16:32,790 --> 00:16:34,826 looking at a blood sample where you 476 00:16:34,848 --> 00:16:36,538 say, yeah, it's in your system, 477 00:16:36,704 --> 00:16:38,906 but we don't know if it actually reached the 478 00:16:38,928 --> 00:16:40,830 point where it should be 479 00:16:40,830 --> 00:16:42,974 doing the job. So again, 480 00:16:43,012 --> 00:16:45,742 these things, I think, open up new ways 481 00:16:45,796 --> 00:16:47,550 and, and trying these 482 00:16:47,620 --> 00:16:49,626 these new methods of sample 483 00:16:49,658 --> 00:16:51,982 collection. And then, of course, 484 00:16:52,036 --> 00:16:54,946 having the perfect tool that analyzes these 485 00:16:54,968 --> 00:16:56,386 samples. And again, the 486 00:16:56,408 --> 00:16:58,674 fantastic low 487 00:16:58,712 --> 00:17:00,766 volume requirement of Olink 488 00:17:00,798 --> 00:17:02,946 has, for us, been this 489 00:17:02,968 --> 00:17:04,770 perfect match. 490 00:17:04,770 --> 00:17:06,806 So we're super happy that we have a 491 00:17:06,828 --> 00:17:08,966 tool that we can test these ideas and we 492 00:17:08,988 --> 00:17:10,982 can demonstrate it's actually 493 00:17:11,036 --> 00:17:13,690 feasible. 494 00:17:13,690 --> 00:17:15,926 To return to what you're excited about in 495 00:17:15,948 --> 00:17:17,986 terms of these longitudinal studies, 496 00:17:18,098 --> 00:17:20,082 have you had much interaction with the UK 497 00:17:20,146 --> 00:17:22,954 Biobank in terms of samples at 498 00:17:22,992 --> 00:17:24,682 scale? I guess you don't have to worry about 499 00:17:24,736 --> 00:17:26,838 sort of the dried blood spot collection. 500 00:17:26,934 --> 00:17:28,954 I mean, that's really promising, but 501 00:17:28,992 --> 00:17:30,762 here it is. We have a huge 502 00:17:30,816 --> 00:17:32,458 data set. Have you been involved much with 503 00:17:32,464 --> 00:17:34,790 the UK Biobank? 504 00:17:34,790 --> 00:17:36,846 Indirectly, yes. I mean, I've been 505 00:17:36,868 --> 00:17:38,878 talking to Chris Whelan and others. And of 506 00:17:38,884 --> 00:17:40,206 course, when 507 00:17:40,308 --> 00:17:42,866 Karsten Suhre, Mark McCarthy, and I 508 00:17:42,888 --> 00:17:44,786 started to write this review in 509 00:17:44,808 --> 00:17:46,722 Nature Genetics a couple of years back, 510 00:17:46,856 --> 00:17:48,626 we thought of UK Biobank as the 511 00:17:48,648 --> 00:17:50,830 audience, 512 00:17:50,830 --> 00:17:52,440 513 00:17:52,550 --> 00:17:54,274 especially Mark 514 00:17:54,392 --> 00:17:56,920 McCarthy who I consider my mentor. 515 00:17:56,920 --> 00:17:58,726 He was in Stockholm and I talked to him 516 00:17:58,748 --> 00:18:00,994 and said, "Mark, you're doing this fantastic 517 00:18:01,042 --> 00:18:03,234 work. And I think proteomics like Karsten 518 00:18:03,282 --> 00:18:05,126 Suhre has shown, is a perfect match with 519 00:18:05,148 --> 00:18:07,766 genetics. Can't we write up something as 520 00:18:07,868 --> 00:18:09,798 bringing different perspectives together 521 00:18:09,884 --> 00:18:11,480 into one piece of 522 00:18:11,484 --> 00:18:13,466 information?" And that's sort of how this 523 00:18:13,488 --> 00:18:15,322 whole idea started. We actually called 524 00:18:15,376 --> 00:18:17,242 up Karsten and said, "Karsten, we have this idea, 525 00:18:17,296 --> 00:18:19,994 do you want to join?" And this is sort of 526 00:18:20,192 --> 00:18:22,526 where we joined forces. I learned so much 527 00:18:22,548 --> 00:18:24,574 about genetics, and others learned about 528 00:18:24,612 --> 00:18:26,846 proteomics. So I think that 529 00:18:26,868 --> 00:18:28,920 sort of was, of course, 530 00:18:28,920 --> 00:18:30,666 the dream 531 00:18:30,778 --> 00:18:32,842 coming true as writing 532 00:18:32,906 --> 00:18:34,402 something that adds value, but learning 533 00:18:34,456 --> 00:18:36,946 something at the same time. And then, of 534 00:18:36,968 --> 00:18:38,870 course, UK Biobank 535 00:18:38,870 --> 00:18:40,658 being, as has been shown, a 536 00:18:40,664 --> 00:18:42,546 fantastic study 537 00:18:42,648 --> 00:18:44,946 now being powered by all these 538 00:18:44,968 --> 00:18:46,740 new data that is coming out. 539 00:18:46,740 --> 00:18:48,658 But, yeah, again, 540 00:18:48,824 --> 00:18:50,998 it's often a one timepoint 541 00:18:51,084 --> 00:18:53,366 picture, but we want to create a movie of 542 00:18:53,388 --> 00:18:55,686 our lives, right? And the movie tells the 543 00:18:55,708 --> 00:18:57,158 story much better. And we should probably 544 00:18:57,244 --> 00:18:59,630 just explain Mark McCarthy, although I don't 545 00:18:59,650 --> 00:19:01,674 think he needs an introduction. He's such 546 00:19:01,712 --> 00:19:03,866 a well-known figure in 547 00:19:03,888 --> 00:19:05,466 our world, certainly, but he's at 548 00:19:05,488 --> 00:19:07,434 Genentech, of course, but he's one of these 549 00:19:07,472 --> 00:19:09,562 geneticists that has 550 00:19:09,616 --> 00:19:11,566 crossed over into industry. And 551 00:19:11,668 --> 00:19:13,520 just anything he 552 00:19:13,520 --> 00:19:15,854 focuses on I like to 553 00:19:15,892 --> 00:19:17,582 keep an eye on, because it 554 00:19:17,636 --> 00:19:19,980 moves and shakes. He was at 555 00:19:19,980 --> 00:19:21,374 the International Congress of Human 556 00:19:21,412 --> 00:19:23,906 Genetics, and so involved in the 557 00:19:23,928 --> 00:19:25,906 leadership, talking about how to 558 00:19:25,928 --> 00:19:27,920 increase diversity in genetics. And 559 00:19:27,920 --> 00:19:29,874 I love 560 00:19:29,912 --> 00:19:31,874 that Nature Genetics paper. So I just wanted 561 00:19:31,912 --> 00:19:33,950 to say, "Karsten, 562 00:19:34,030 --> 00:19:36,854 you, and Mark M, it's just 563 00:19:36,892 --> 00:19:38,902 such a pleasure to have 564 00:19:39,036 --> 00:19:41,346 you even talking about our technology. It's 565 00:19:41,378 --> 00:19:43,790 very exciting." 566 00:19:43,790 --> 00:19:45,878 And I think, 567 00:19:45,964 --> 00:19:47,766 of course, we wanted to be 568 00:19:47,788 --> 00:19:49,746 as agnostic and fair as 569 00:19:49,788 --> 00:19:51,910 possible, because I think 570 00:19:51,910 --> 00:19:53,846 every technology has its pros 571 00:19:53,878 --> 00:19:55,834 and cons, and I think it's up 572 00:19:55,872 --> 00:19:57,498 to everyone to make a decision what is the 573 00:19:57,504 --> 00:19:59,978 best fit for the situation. 574 00:20:00,064 --> 00:20:02,718 Absolutely. But I guess 575 00:20:02,804 --> 00:20:04,190 coming back to your question about 576 00:20:04,260 --> 00:20:06,810 longitudinal studies, 577 00:20:06,850 --> 00:20:08,862 which we've been also doing 578 00:20:08,996 --> 00:20:10,782 locally, led by 579 00:20:10,836 --> 00:20:12,782 Mathias Uhlén, and we've been working with 580 00:20:12,836 --> 00:20:14,554 Jochen Schwenk from the SCALLOP 581 00:20:14,602 --> 00:20:16,858 cohort. You know, of course, that 582 00:20:17,044 --> 00:20:19,490 that is when it all sort of comes to life, 583 00:20:19,560 --> 00:20:21,986 right? When you see a signature, you can 584 00:20:22,008 --> 00:20:24,770 understand stability, you can understand 585 00:20:24,770 --> 00:20:26,946 that a person has had an infection, 586 00:20:27,058 --> 00:20:29,382 things go up, things go down, but someone 587 00:20:29,436 --> 00:20:31,986 loses weight, things change. So that's 588 00:20:32,018 --> 00:20:34,570 when the information actually becomes 589 00:20:34,570 --> 00:20:36,486 clearer. And that's a 590 00:20:36,508 --> 00:20:38,934 fascinating thing: to be able to look 591 00:20:38,972 --> 00:20:40,946 at this real time biology. 592 00:20:41,138 --> 00:20:43,674 I appreciate you talking about this 593 00:20:43,712 --> 00:20:45,754 review paper for the audience. 594 00:20:45,792 --> 00:20:47,866 The paper I believe you're talking about is 595 00:20:47,888 --> 00:20:49,462 "Genetics meets Proteomics: 596 00:20:49,526 --> 00:20:51,930 Perspectives for Large Population-based 597 00:20:52,000 --> 00:20:54,846 Studies." It was in Nature Review Genetics in 598 00:20:54,868 --> 00:20:56,800 January 2021. 599 00:20:56,800 --> 00:20:58,846 I'm trying to remember a 600 00:20:58,868 --> 00:21:00,926 different Karsten Suhre review. I 601 00:21:00,948 --> 00:21:02,890 think you're talking about maybe one from 602 00:21:02,890 --> 00:21:04,958 2017 603 00:21:05,054 --> 00:21:07,820 or 2019. At any rate, 604 00:21:07,820 --> 00:21:09,714 the ability to 605 00:21:09,752 --> 00:21:11,794 monitor real time health as 606 00:21:11,832 --> 00:21:13,746 people transition from a state of 607 00:21:13,768 --> 00:21:15,662 health to one of disease. 608 00:21:15,806 --> 00:21:17,686 I finished a book recently, "The Age of 609 00:21:17,708 --> 00:21:19,554 Scientific Wellness," from Leroy Hood 610 00:21:19,602 --> 00:21:21,814 and Nathan Price, 611 00:21:21,932 --> 00:21:23,746 and it talked about these disease 612 00:21:23,858 --> 00:21:25,686 transitions, where if 613 00:21:25,708 --> 00:21:27,634 somebody's healthy, they don't have 614 00:21:27,692 --> 00:21:29,754 symptoms, but something's happening 615 00:21:29,872 --> 00:21:31,626 in the body, something's happening with 616 00:21:31,648 --> 00:21:33,386 their metabolism, something's happening with 617 00:21:33,408 --> 00:21:35,290 their metagenomics, something's happening 618 00:21:35,360 --> 00:21:37,146 with their proteomics and the 619 00:21:37,168 --> 00:21:39,834 circulation. And that is just 620 00:21:39,872 --> 00:21:41,306 this fascinating thing because you're 621 00:21:41,338 --> 00:21:43,342 talking about wellness, right? We need to be 622 00:21:43,396 --> 00:21:45,902 sampling "well" people. And I think 623 00:21:46,036 --> 00:21:48,026 the UK Biobank gives this unique 624 00:21:48,058 --> 00:21:50,574 perspective. I'd like to hear your 625 00:21:50,612 --> 00:21:52,920 perspective on that. 626 00:21:52,920 --> 00:21:54,846 Yeah, of course, I mean, 627 00:21:54,948 --> 00:21:56,914 UK Biobank offers, as far as I 628 00:21:56,952 --> 00:21:58,958 understand, really a range 629 00:21:58,974 --> 00:22:00,910 of phenotypes. I assume 630 00:22:01,070 --> 00:22:03,860 some involvement was in selecting particular 631 00:22:03,860 --> 00:22:05,686 sort of 632 00:22:05,708 --> 00:22:07,574 disease groups and enriching them for those 633 00:22:07,612 --> 00:22:09,462 that are maybe more prevalent than others. 634 00:22:09,516 --> 00:22:11,222 But just to have that 635 00:22:11,276 --> 00:22:13,526 breadth is really amazing 636 00:22:13,628 --> 00:22:15,442 because often you're limited 637 00:22:15,506 --> 00:22:17,370 to 638 00:22:17,370 --> 00:22:19,634 certain sample collections. 639 00:22:19,682 --> 00:22:21,894 And maybe I 640 00:22:21,932 --> 00:22:23,606 take another sort of, open another bracket 641 00:22:23,638 --> 00:22:25,322 and take a little detour here. But again, 642 00:22:25,376 --> 00:22:27,254 when you do this dried blood spot random 643 00:22:27,302 --> 00:22:29,802 sampling that we did, you include everyone. 644 00:22:29,936 --> 00:22:31,706 You don't include only the ones that are 645 00:22:31,728 --> 00:22:33,466 sick, and they only come when they're sick. 646 00:22:33,498 --> 00:22:35,694 So, you know, okay, CRP [C-reactive protein] and all the other 647 00:22:35,732 --> 00:22:37,822 friends, they're all already up, right? 648 00:22:37,956 --> 00:22:39,614 But we want - so how do you get 649 00:22:39,652 --> 00:22:41,758 that cross sectional, that 650 00:22:41,844 --> 00:22:43,838 true sort of population-based 651 00:22:43,924 --> 00:22:45,970 variance? And I think that's only possible 652 00:22:46,040 --> 00:22:48,674 in a coordinated way, like UK Biobank did. 653 00:22:48,712 --> 00:22:50,990 And there are other biobanks that 654 00:22:50,990 --> 00:22:52,866 all of us in the U.S. and 655 00:22:52,888 --> 00:22:54,354 others are trying to do similar 656 00:22:54,472 --> 00:22:56,886 things. That when you learn this 657 00:22:56,908 --> 00:22:58,934 is the human variability with all the 658 00:22:58,972 --> 00:23:00,758 genetics, with lifestyle, with 659 00:23:00,844 --> 00:23:02,802 social economic factors 660 00:23:02,866 --> 00:23:04,870 influencing 661 00:23:04,870 --> 00:23:06,902 who you are on a molecular level. 662 00:23:07,036 --> 00:23:09,082 So yeah, fantastic. And then having 663 00:23:09,136 --> 00:23:11,962 proteomics in that play is of course 664 00:23:12,016 --> 00:23:14,390 something I get particularly excited about. 665 00:23:14,390 --> 00:23:16,442 And it's this 666 00:23:16,496 --> 00:23:18,746 combination - I'm sorry, go 667 00:23:18,768 --> 00:23:20,970 ahead. I'm sorry, Dale, 668 00:23:20,970 --> 00:23:22,750 go first. 669 00:23:22,750 --> 00:23:24,926 Oh no, Sarantis, this is your show. Go right 670 00:23:24,948 --> 00:23:26,840 ahead, please. 671 00:23:26,840 --> 00:23:28,974 Thank you. Now I'm talking 672 00:23:28,990 --> 00:23:30,542 about how you mentioned about 673 00:23:30,596 --> 00:23:32,414 proteomics, genomics and 674 00:23:32,452 --> 00:23:34,906 disease, and there's 675 00:23:34,938 --> 00:23:36,546 a preprint with Anders Malarstig now 676 00:23:36,568 --> 00:23:38,466 that recently came out, and they're going to 677 00:23:38,488 --> 00:23:40,718 be soon published - finger crossed - with Olink. 678 00:23:40,814 --> 00:23:42,114 Would you like to share a little bit 679 00:23:42,152 --> 00:23:44,930 information about the research there 680 00:23:45,000 --> 00:23:46,786 and the cohort that you use and what's your 681 00:23:46,808 --> 00:23:48,754 main findings? Because it's exciting 682 00:23:48,802 --> 00:23:50,866 to use the circulating proteome 683 00:23:50,898 --> 00:23:52,838 to identify prognosis [biomarkers] for 684 00:23:52,924 --> 00:23:54,786 breast cancer, early prognosis for breast 685 00:23:54,818 --> 00:23:56,966 cancer. I'm happy to hear a little bit more 686 00:23:56,988 --> 00:23:58,750 from you, actually. 687 00:23:58,750 --> 00:24:00,874 Yeah, so 688 00:24:00,912 --> 00:24:02,986 we're talking about the KARMA Cohort, which 689 00:24:03,008 --> 00:24:05,814 is a Swedish breast cancer population 690 00:24:05,862 --> 00:24:07,914 cohort that invites all women 691 00:24:07,952 --> 00:24:09,606 in Sweden undergoing mammography 692 00:24:09,638 --> 00:24:11,706 screening to participate. This is 693 00:24:11,728 --> 00:24:13,690 spearheaded by Per Hall and 694 00:24:13,760 --> 00:24:15,410 Kamila Czene at Karolinska 695 00:24:15,430 --> 00:24:17,646 Institute. And with both I've been 696 00:24:17,668 --> 00:24:19,374 collaborating already for quite some 697 00:24:19,412 --> 00:24:21,854 time and we recently got 698 00:24:21,892 --> 00:24:23,854 some funding to continue our collaborations 699 00:24:23,902 --> 00:24:25,202 and then brought in also 700 00:24:25,256 --> 00:24:27,790 Olink data that we generated 701 00:24:27,950 --> 00:24:29,610 in the lab 702 00:24:29,610 --> 00:24:31,646 to look at breast 703 00:24:31,678 --> 00:24:33,906 cancer risk. So in addition to this 704 00:24:33,928 --> 00:24:35,578 paper that you mentioned, Sarantis, there's 705 00:24:35,614 --> 00:24:37,862 also another one that's been 706 00:24:37,916 --> 00:24:39,990 circling around now 707 00:24:39,990 --> 00:24:41,250 where we wanted to primarily 708 00:24:41,330 --> 00:24:43,842 identify - Can we use proteins 709 00:24:43,906 --> 00:24:45,846 to predict short term risk of 710 00:24:45,868 --> 00:24:47,766 breast cancer? Genetics can do that on 711 00:24:47,788 --> 00:24:49,790 a more longer period 712 00:24:49,810 --> 00:24:51,894 of time, but can proteins 713 00:24:51,942 --> 00:24:53,818 add something to it? And then, of 714 00:24:53,824 --> 00:24:55,940 course, with Anderson and 715 00:24:55,940 --> 00:24:57,494 the leader and the SCALLOP 716 00:24:57,542 --> 00:24:59,754 consortium, and we want to also to bring 717 00:24:59,792 --> 00:25:01,930 also genetics into this. And 718 00:25:01,930 --> 00:25:03,886 I'm not a geneticist. So for 719 00:25:03,908 --> 00:25:05,790 me, again, it's always fascinating to see 720 00:25:05,860 --> 00:25:07,790 proteomics data in action. 721 00:25:07,790 --> 00:25:09,920 722 00:25:09,920 --> 00:25:11,854 Which is what 723 00:25:11,892 --> 00:25:13,626 it makes me most proud 724 00:25:13,658 --> 00:25:15,760 because I think 725 00:25:15,760 --> 00:25:17,886 what it is exciting to 726 00:25:17,908 --> 00:25:19,842 learn when other datas inform you about 727 00:25:19,896 --> 00:25:21,714 your own data and when others take the data 728 00:25:21,752 --> 00:25:23,906 that you generate or that you 729 00:25:23,928 --> 00:25:25,886 know more about and they tell you new 730 00:25:25,928 --> 00:25:27,890 stories. And then, of course, 731 00:25:27,890 --> 00:25:29,930 732 00:25:29,930 --> 00:25:31,990 the KARMA cohort is really a population-based 733 00:25:32,060 --> 00:25:34,406 code and it's really unique in a sense that 734 00:25:34,428 --> 00:25:36,710 we're not only looking at 735 00:25:36,710 --> 00:25:38,980 breast cancer cases, 736 00:25:38,980 --> 00:25:40,474 or the study could 737 00:25:40,512 --> 00:25:42,966 continuously collect sample 738 00:25:43,158 --> 00:25:45,338 patient information or personal information, 739 00:25:45,504 --> 00:25:47,786 and eventually some of these persons will 740 00:25:47,808 --> 00:25:49,994 become patients. And luckily, then 741 00:25:50,032 --> 00:25:51,430 we would have, let's say, a blood 742 00:25:51,478 --> 00:25:53,746 sample from the last time when the patient 743 00:25:53,798 --> 00:25:55,534 was still a person, so to speak, when you 744 00:25:55,572 --> 00:25:57,886 think about these two categories. So we can 745 00:25:57,908 --> 00:25:59,966 go back in time and see are there any 746 00:26:00,148 --> 00:26:02,870 things in the 747 00:26:02,870 --> 00:26:04,610 prior history that could 748 00:26:04,680 --> 00:26:06,914 lead towards okay, you 749 00:26:06,952 --> 00:26:08,402 are actually on a much different 750 00:26:08,456 --> 00:26:10,382 trajectory than the remaining 751 00:26:10,446 --> 00:26:12,994 individuals, so that's all that 752 00:26:13,112 --> 00:26:15,630 people have different lives. We know drugs 753 00:26:15,710 --> 00:26:17,786 played a role, pre- post-menopausal 754 00:26:17,838 --> 00:26:19,770 plays a role, 755 00:26:19,770 --> 00:26:21,986 hormonal replacement therapy plays a role. 756 00:26:22,018 --> 00:26:24,674 So lots of things happen. But then genetics 757 00:26:24,722 --> 00:26:26,774 can tell you an unbiased story about all 758 00:26:26,812 --> 00:26:28,746 these phenotypes and that, 759 00:26:28,768 --> 00:26:30,858 again, gives a new angle to 760 00:26:30,944 --> 00:26:32,922 this whole problem. And in this study, 761 00:26:32,976 --> 00:26:34,166 we used Mendelian 762 00:26:34,198 --> 00:26:36,762 Randomization and found five 763 00:26:36,816 --> 00:26:38,282 interesting proteins that 764 00:26:38,416 --> 00:26:40,954 presumably have a causal role in breast 765 00:26:41,002 --> 00:26:43,566 cancer. And of course, this is the study. 766 00:26:43,748 --> 00:26:45,978 Now, it's only nowadays 767 00:26:46,074 --> 00:26:48,974 600 individuals, but still, it's a 768 00:26:49,012 --> 00:26:51,758 really very fine selection of 769 00:26:51,924 --> 00:26:53,934 samples that could 770 00:26:54,052 --> 00:26:56,978 lead the way. And then again, taking the 771 00:26:56,984 --> 00:26:58,514 road that genetics has 772 00:26:58,632 --> 00:27:00,866 taken, we can use data that exists in 773 00:27:00,888 --> 00:27:02,946 other biobanks and we can sort 774 00:27:02,968 --> 00:27:04,626 of look, do we see the 775 00:27:04,648 --> 00:27:06,886 same associations in these? And 776 00:27:06,908 --> 00:27:08,880 that's, of course, 777 00:27:08,880 --> 00:27:10,786 when multiomics doesn't 778 00:27:10,818 --> 00:27:12,806 become a picture, it becomes a movie, where 779 00:27:12,828 --> 00:27:14,902 we take different [aspects] of these 780 00:27:14,956 --> 00:27:16,810 relationships. 781 00:27:16,810 --> 00:27:18,566 What a great illustration 782 00:27:18,678 --> 00:27:20,860 and analogy. 783 00:27:20,860 --> 00:27:22,858 That's a 784 00:27:22,864 --> 00:27:24,838 great analogy, right? Not a picture. We've 785 00:27:24,854 --> 00:27:26,850 got a movie. 786 00:27:26,850 --> 00:27:28,938 Yeah. And I think that's what 787 00:27:28,944 --> 00:27:30,980 we need, right? 788 00:27:30,980 --> 00:27:32,986 You take 789 00:27:33,008 --> 00:27:35,166 a look at a picture and you interpret so 790 00:27:35,188 --> 00:27:37,966 many things into this, whether you 791 00:27:37,988 --> 00:27:39,662 know something about the painter or 792 00:27:39,716 --> 00:27:41,854 the time when the painting was made. 793 00:27:41,892 --> 00:27:43,938 But if you have a movie, it tells you 794 00:27:44,024 --> 00:27:46,386 much more. It tells you a dynamic that 795 00:27:46,408 --> 00:27:48,658 you cannot really see in a 796 00:27:48,664 --> 00:27:50,594 picture. But anyway, so 797 00:27:50,632 --> 00:27:52,850 again, we had this opportunity, and then, 798 00:27:52,850 --> 00:27:54,630 799 00:27:54,630 --> 00:27:56,940 Asa [Hedman] and Anders [Malarstig] have been really leading 800 00:27:56,940 --> 00:27:58,962 this together with colleagues at Olink 801 00:27:59,026 --> 00:28:01,798 and others, to sort of 802 00:28:01,964 --> 00:28:03,654 find out whether these things we 803 00:28:03,692 --> 00:28:05,862 identified in the Swedish KARMA study also 804 00:28:05,916 --> 00:28:07,446 we can see in the UK Biobank or 805 00:28:07,468 --> 00:28:09,954 in Finngen and it seems so to be the case. 806 00:28:10,012 --> 00:28:12,250 And of course, that gives much more 807 00:28:12,400 --> 00:28:14,842 certainty about that. These are interesting 808 00:28:14,896 --> 00:28:16,730 findings to follow up. And 809 00:28:16,800 --> 00:28:18,634 again, I think what we talked, I guess 810 00:28:18,672 --> 00:28:20,270 before this podcast started, 811 00:28:20,270 --> 00:28:22,906 that then you can 812 00:28:22,928 --> 00:28:24,782 start to develop drugs, and you can see what 813 00:28:24,836 --> 00:28:26,718 actually happens when you give someone a 814 00:28:26,724 --> 00:28:28,778 drug that addresses one of these proteins. 815 00:28:28,874 --> 00:28:30,762 Then you, again, start a new movie, 816 00:28:30,826 --> 00:28:32,750 right? But, 817 00:28:32,750 --> 00:28:34,850 on a different direction. 818 00:28:34,850 --> 00:28:36,674 And again, then use 819 00:28:36,712 --> 00:28:38,594 proteomics to follow and see what 820 00:28:38,632 --> 00:28:40,738 happens. So, 821 00:28:40,830 --> 00:28:42,866 that's fascinating, I think. And so 822 00:28:42,888 --> 00:28:44,750 who does that follow up? 823 00:28:44,750 --> 00:28:46,902 Are you involved in that 824 00:28:46,956 --> 00:28:48,490 kind of 825 00:28:48,490 --> 00:28:49,910 obviously, 826 00:28:49,910 --> 00:28:51,430 827 00:28:51,430 --> 00:28:53,800 proteomics 828 00:28:53,800 --> 00:28:55,878 is your field, 829 00:28:56,044 --> 00:28:58,486 and I just wonder if there's 830 00:28:58,518 --> 00:29:00,590 another 831 00:29:00,590 --> 00:29:02,666 function that takes that to the 832 00:29:02,688 --> 00:29:04,954 clinical trial or to the 833 00:29:04,992 --> 00:29:06,950 test bed to 834 00:29:06,950 --> 00:29:08,618 try out 835 00:29:08,784 --> 00:29:10,974 these drugs that 836 00:29:11,012 --> 00:29:13,930 affect these pathways. 837 00:29:13,930 --> 00:29:15,920 Yeah. 838 00:29:16,030 --> 00:29:18,414 Of course, I guess it would require that 839 00:29:18,452 --> 00:29:20,366 we have the right partners who would have 840 00:29:20,388 --> 00:29:22,506 the libraries to do drug screening on these, 841 00:29:22,548 --> 00:29:24,434 and sort of it's an army of new 842 00:29:24,472 --> 00:29:26,850 things to engage. 843 00:29:26,850 --> 00:29:28,790 But of course, 844 00:29:28,790 --> 00:29:30,914 primarily to see that 845 00:29:30,952 --> 00:29:32,594 what we do in these 846 00:29:32,632 --> 00:29:34,994 studies has a value and then 847 00:29:35,032 --> 00:29:37,186 again, translate it back to functional 848 00:29:37,218 --> 00:29:39,690 studies, which, again, is something 849 00:29:39,690 --> 00:29:41,974 I think will also happen in the next couple 850 00:29:42,012 --> 00:29:44,214 of years, is taking all these big 851 00:29:44,332 --> 00:29:46,886 biobank screenings back into some sort of 852 00:29:46,908 --> 00:29:48,858 functional studies to see, okay, is it 853 00:29:48,864 --> 00:29:50,906 really the molecule? Is it really the 854 00:29:50,928 --> 00:29:52,426 phenotype? Is it 855 00:29:52,448 --> 00:29:54,650 really the drug or the 856 00:29:54,720 --> 00:29:56,986 lifestyle effect? And 857 00:29:57,088 --> 00:29:59,626 that's going to be sort of looping back 858 00:29:59,808 --> 00:30:01,642 where it started, from cellular 859 00:30:01,706 --> 00:30:03,490 studies into 860 00:30:03,490 --> 00:30:05,822 systemic studies, and then back into the - 861 00:30:05,956 --> 00:30:07,806 Wouldn't it be amazing to 862 00:30:07,828 --> 00:30:09,822 find a lifestyle effect that 863 00:30:09,876 --> 00:30:11,854 we never thought might have 864 00:30:11,892 --> 00:30:13,990 an impact? 865 00:30:13,990 --> 00:30:15,986 Yeah. Having the tools to 866 00:30:16,008 --> 00:30:18,894 be able to start parsing these things. It's 867 00:30:18,942 --> 00:30:20,810 fascinating. 868 00:30:22,810 --> 00:30:24,840 869 00:30:26,840 --> 00:30:28,742 To talk a little bit about the 870 00:30:28,796 --> 00:30:30,354 study itself, right? This Nature 871 00:30:30,402 --> 00:30:32,438 Preprint looked at 300 872 00:30:32,524 --> 00:30:34,694 individuals from this mammography study who 873 00:30:34,732 --> 00:30:36,466 had breast cancer 874 00:30:36,578 --> 00:30:38,530 diagnosed over those two years, that 875 00:30:38,540 --> 00:30:40,666 they took a look at them, and then you 876 00:30:40,688 --> 00:30:42,986 matched it with 300 normal individuals from 877 00:30:43,008 --> 00:30:45,254 that same study. You had genotype 878 00:30:45,302 --> 00:30:47,194 information, right? And the 879 00:30:47,232 --> 00:30:49,914 genotypes combined with the 880 00:30:49,930 --> 00:30:51,886 Explore 3000 [platform] in 881 00:30:51,908 --> 00:30:53,486 terms of the 2900 882 00:30:53,588 --> 00:30:55,422 proteins. So you had 600 883 00:30:55,476 --> 00:30:57,582 individuals, 2900 884 00:30:57,636 --> 00:30:59,434 proteins, and then you discovered 885 00:30:59,482 --> 00:31:01,274 800 pQTLs [protein Quantitative Trait Loci] 886 00:31:01,402 --> 00:31:03,322 and controlling 887 00:31:03,386 --> 00:31:05,906 737 proteins. And I 888 00:31:05,928 --> 00:31:07,666 thought that was fascinating, right? That we 889 00:31:07,688 --> 00:31:09,570 have genetic control of 890 00:31:09,640 --> 00:31:11,634 737 proteins that can 891 00:31:11,672 --> 00:31:13,102 identify the variants 892 00:31:13,166 --> 00:31:15,974 pQTLs, and then you can drill down 893 00:31:16,012 --> 00:31:18,450 and get five likely 894 00:31:18,530 --> 00:31:20,966 causative proteins of breast cancer. Do I 895 00:31:20,988 --> 00:31:22,902 understand that correctly? That these 896 00:31:22,956 --> 00:31:24,690 five proteins you identified 897 00:31:24,850 --> 00:31:26,822 were previously not 898 00:31:26,876 --> 00:31:28,954 investigated or investigated with 899 00:31:28,992 --> 00:31:30,694 certain sort of weak associations 900 00:31:30,742 --> 00:31:32,930 of breast cancer? But the take home 901 00:31:32,992 --> 00:31:34,746 message is that these five proteins were 902 00:31:34,768 --> 00:31:36,930 new discoveries. 903 00:31:36,930 --> 00:31:38,794 Yeah, that's our 904 00:31:38,832 --> 00:31:40,726 understanding. I mean, of course, 905 00:31:40,768 --> 00:31:42,814 maybe someone else has already figured this 906 00:31:42,852 --> 00:31:44,494 out, but not told the public about 907 00:31:44,532 --> 00:31:46,878 it. But I 908 00:31:46,884 --> 00:31:48,850 think 909 00:31:48,850 --> 00:31:50,734 an important aspect is also to say 910 00:31:50,772 --> 00:31:52,922 that these 300 cases, 911 00:31:52,986 --> 00:31:54,786 they were not cases at the time of 912 00:31:54,808 --> 00:31:56,914 sampling, they were future cases. So when 913 00:31:56,952 --> 00:31:58,882 they were actually sampled, they were still 914 00:31:59,016 --> 00:32:01,910 considered persons, not patients. 915 00:32:01,910 --> 00:32:03,530 So that's, again, also 916 00:32:03,530 --> 00:32:05,790 to understand, 917 00:32:05,790 --> 00:32:07,894 then we have this list 918 00:32:07,932 --> 00:32:09,974 of proteins that all tell 919 00:32:10,012 --> 00:32:12,386 different stories. And I think it's 920 00:32:12,418 --> 00:32:14,854 fascinating to be sort of in your 921 00:32:14,892 --> 00:32:16,806 mind, thinking about what actually the 922 00:32:16,828 --> 00:32:18,886 role is. But what we need now is, of 923 00:32:18,908 --> 00:32:20,954 course, the hard data that tells us this 924 00:32:20,992 --> 00:32:22,794 is true, or this is 925 00:32:22,832 --> 00:32:24,900 actually the opposite. 926 00:32:24,900 --> 00:32:26,986 And I guess to back up 927 00:32:27,008 --> 00:32:29,386 to the original KARMA study, which was out 928 00:32:29,408 --> 00:32:31,342 of KTH. There were some 929 00:32:31,396 --> 00:32:33,614 70,000 women from Karolinska who 930 00:32:33,652 --> 00:32:35,886 volunteered. From Karolinska. 931 00:32:35,908 --> 00:32:37,550 Okay. 70,000 women, 932 00:32:37,620 --> 00:32:39,806 though, over a couple of years. Is 933 00:32:39,828 --> 00:32:41,902 that correct? I think about 934 00:32:41,956 --> 00:32:43,374 the effort 935 00:32:43,422 --> 00:32:45,790 involved. 936 00:32:45,790 --> 00:32:47,714 Yeah. This is the nice thing about 937 00:32:47,752 --> 00:32:49,730 doing science. In Sweden - 938 00:32:49,730 --> 00:32:51,666 I originally come from 939 00:32:51,688 --> 00:32:53,938 Germany - it's a different 940 00:32:54,024 --> 00:32:56,546 system. But in Sweden, 941 00:32:56,738 --> 00:32:58,742 maybe because of the Nobel Prize, maybe 942 00:32:58,796 --> 00:33:00,982 because of the public interest 943 00:33:01,036 --> 00:33:03,282 in science, there's much easier 944 00:33:03,346 --> 00:33:05,814 engagement. And in 945 00:33:05,852 --> 00:33:07,974 women, I think, in other countries, have 946 00:33:08,012 --> 00:33:10,042 this regular sort of health checkup. So 947 00:33:10,096 --> 00:33:11,978 there's this mammography screening program, 948 00:33:12,064 --> 00:33:14,090 and then you get basically asked, "Do 949 00:33:14,208 --> 00:33:16,774 you want to participate?" And then Per Hall 950 00:33:16,822 --> 00:33:18,906 and his colleagues do the 951 00:33:18,928 --> 00:33:20,390 magic and keep people 952 00:33:20,560 --> 00:33:22,510 engaged, and people 953 00:33:22,580 --> 00:33:24,942 follow, which 954 00:33:25,076 --> 00:33:27,966 is super. And 955 00:33:27,988 --> 00:33:29,994 then you mentioned briefly 956 00:33:30,042 --> 00:33:32,734 the power of replicating these 957 00:33:32,772 --> 00:33:34,974 results, because that, 958 00:33:35,012 --> 00:33:36,866 I think, is an important dimension of this 959 00:33:36,888 --> 00:33:38,754 paper, in that it wasn't just a 960 00:33:38,792 --> 00:33:40,290 single finding in a particular 961 00:33:40,360 --> 00:33:42,866 population that you were able 962 00:33:42,888 --> 00:33:44,946 to find. We're actually able then to go 963 00:33:44,968 --> 00:33:46,946 back to was it? Finngen and the UK 964 00:33:46,978 --> 00:33:48,694 biobank? And then look 965 00:33:48,732 --> 00:33:50,854 at the genotypes, look at the 966 00:33:50,892 --> 00:33:52,966 protein levels, and then being able to 967 00:33:52,988 --> 00:33:54,946 actually show, yes, this connection holds 968 00:33:54,978 --> 00:33:56,422 up. I think that's pretty 969 00:33:56,476 --> 00:33:58,970 significant. 970 00:33:58,970 --> 00:34:00,860 Could you comment on that? 971 00:34:02,860 --> 00:34:04,854 Again, 972 00:34:04,972 --> 00:34:06,886 this work was spearheaded by Asa Hedman, 973 00:34:06,918 --> 00:34:08,970 but again, what I see is 974 00:34:09,040 --> 00:34:11,674 you have, let's say you create this 975 00:34:11,712 --> 00:34:13,726 currency, let's say the 976 00:34:13,748 --> 00:34:15,870 pQTLs. This is a currency you can 977 00:34:15,940 --> 00:34:17,840 go and you can pay in other 978 00:34:17,840 --> 00:34:19,534 countries or in other 979 00:34:19,572 --> 00:34:21,486 biobanks. You can use that currency to 980 00:34:21,508 --> 00:34:23,726 exchange information. And 981 00:34:23,828 --> 00:34:25,794 this is, I guess, what genetics has 982 00:34:25,832 --> 00:34:27,650 really enabled us to do. And now 983 00:34:27,720 --> 00:34:29,874 proteomics is learning how it can do 984 00:34:29,912 --> 00:34:31,522 it. We have different 985 00:34:31,576 --> 00:34:33,042 technologies, they may have different 986 00:34:33,096 --> 00:34:35,682 outcomes, different information. But again, 987 00:34:35,736 --> 00:34:37,666 you can anchor it on the genetics. You can 988 00:34:37,688 --> 00:34:39,930 use the pQTLS, you can 989 00:34:39,930 --> 00:34:41,938 use them as instruments in mendelian 990 00:34:41,954 --> 00:34:43,878 randomization to exchange this information. 991 00:34:43,964 --> 00:34:45,290 And that's 992 00:34:45,290 --> 00:34:47,938 amazing. Yeah. Here it is. You're 993 00:34:47,954 --> 00:34:49,446 talking about empowering proteomics with 994 00:34:49,468 --> 00:34:51,930 genomics, right? 995 00:34:51,930 --> 00:34:53,706 Turning it around instead of coming at, I 996 00:34:53,728 --> 00:34:55,226 mean, I come from a genomics background. So 997 00:34:55,248 --> 00:34:57,238 I think of it in terms of proteomics adding 998 00:34:57,254 --> 00:34:58,906 to the genomics. Here it is. You come from 999 00:34:58,928 --> 00:35:00,634 the proteomics background, and it's the 1000 00:35:00,672 --> 00:35:02,730 genetics that is really enriching 1001 00:35:02,810 --> 00:35:04,666 the findings. And I think that's 1002 00:35:04,698 --> 00:35:06,874 great. Jochen, I have a basic 1003 00:35:06,922 --> 00:35:08,814 question, and it's very basic. We 1004 00:35:08,852 --> 00:35:10,954 mentioned that one of the factors 1005 00:35:11,002 --> 00:35:13,054 could be lifestyle, like 1006 00:35:13,092 --> 00:35:15,502 environment, but also could be the hormones. 1007 00:35:15,646 --> 00:35:17,982 Do you have, let's say, relatives, 1008 00:35:18,046 --> 00:35:20,654 like mother, sister, or twins 1009 00:35:20,782 --> 00:35:22,978 that you can control in this cohort? I 1010 00:35:22,984 --> 00:35:24,306 imagine there will be also some twins that 1011 00:35:24,328 --> 00:35:26,330 you can, let's say, somehow 1012 00:35:26,350 --> 00:35:28,978 discriminate and identify the genetic 1013 00:35:28,994 --> 00:35:30,998 background versus the environment. This 1014 00:35:31,004 --> 00:35:33,846 is the first part of my 1015 00:35:33,868 --> 00:35:35,846 question. The second part is for sure, 1016 00:35:35,868 --> 00:35:37,254 you would check, like, post- and 1017 00:35:37,292 --> 00:35:39,274 pre- menopausal. Then have you seen 1018 00:35:39,312 --> 00:35:41,878 differences? What's 1019 00:35:41,894 --> 00:35:42,986 your feedback on that? What's your 1020 00:35:43,008 --> 00:35:44,346 experience around this type of 1021 00:35:44,368 --> 00:35:46,830 observations? 1022 00:35:46,830 --> 00:35:48,522 Yeah, I mean, we have had 1023 00:35:48,576 --> 00:35:50,682 previous studies that we published using 1024 00:35:50,736 --> 00:35:52,974 other own technologies that we 1025 00:35:53,092 --> 00:35:55,998 used 510 years ago, where we 1026 00:35:56,084 --> 00:35:58,906 specifically looked at hormonal replacement 1027 00:35:58,938 --> 00:36:00,798 therapy as one of the factors which, 1028 00:36:00,884 --> 00:36:02,834 to our surprise, had really a long 1029 00:36:02,872 --> 00:36:04,690 lasting effect on the women's 1030 00:36:04,690 --> 00:36:06,562 proteome, which 1031 00:36:06,616 --> 00:36:08,594 again, is really 1032 00:36:08,712 --> 00:36:10,690 quite significant. 1033 00:36:10,870 --> 00:36:12,914 And then, of course, that pre- and 1034 00:36:12,952 --> 00:36:14,638 post-menopausal breast cancer. 1035 00:36:14,734 --> 00:36:16,386 Again, this is something I've been 1036 00:36:16,408 --> 00:36:17,894 learning from my colleagues that I work 1037 00:36:17,932 --> 00:36:19,870 with, is very different. 1038 00:36:19,870 --> 00:36:21,718 Then, of course, you need to 1039 00:36:21,724 --> 00:36:23,846 disentangle. So, 1040 00:36:23,948 --> 00:36:25,850 I'm sorry, I want to click 1041 00:36:25,850 --> 00:36:27,922 back to what you just said about 1042 00:36:28,076 --> 00:36:30,326 hormone replacement therapy having a lasting 1043 00:36:30,358 --> 00:36:32,840 effect on the proteome. 1044 00:36:32,840 --> 00:36:34,730 What do you mean by that? 1045 00:36:34,800 --> 00:36:36,966 Meaning that it shifts 1046 00:36:36,998 --> 00:36:38,874 the proteome? But what about the 1047 00:36:38,912 --> 00:36:40,894 risk, the cancer risk, right. 1048 00:36:40,932 --> 00:36:42,910 Because certainly the women's health 1049 00:36:42,980 --> 00:36:44,830 study here in the U.S. 1050 00:36:44,900 --> 00:36:46,942 had led to some concerns around 1051 00:36:46,996 --> 00:36:48,930 that. I'm just curious 1052 00:36:48,930 --> 00:36:50,974 if that's part 1053 00:36:51,012 --> 00:36:53,818 of the impact on the proteome, 1054 00:36:53,914 --> 00:36:55,906 do you think? Or maybe what we 1055 00:36:55,928 --> 00:36:57,858 found in this other study is that we 1056 00:36:57,864 --> 00:36:59,842 had a subset of women that really 1057 00:36:59,896 --> 00:37:01,982 we could sort of see that previous 1058 00:37:02,046 --> 00:37:04,914 use of hormones had 1059 00:37:04,952 --> 00:37:06,850 a significant change 1060 00:37:06,920 --> 00:37:08,902 in their proteome and also increased their 1061 00:37:08,956 --> 00:37:10,674 future risk that they were developing breast 1062 00:37:10,722 --> 00:37:12,870 cancer. So, of course, this really 1063 00:37:12,940 --> 00:37:14,838 sort of showed up. But it's a 1064 00:37:14,844 --> 00:37:16,982 small subset of all the women 1065 00:37:17,036 --> 00:37:19,760 that we tested. 1066 00:37:21,770 --> 00:37:23,830 It's a great thing to follow up. 1067 00:37:25,850 --> 00:37:27,754 Of course, we still need to understand is 1068 00:37:27,792 --> 00:37:29,954 what is the effect of taking hormones? 1069 00:37:30,102 --> 00:37:32,890 Do you actually have remodeling of 1070 00:37:32,890 --> 00:37:34,718 some 1071 00:37:34,804 --> 00:37:36,970 reproductive pathways that 1072 00:37:36,970 --> 00:37:38,958 constantly do something? 1073 00:37:39,044 --> 00:37:41,098 And if they get sort of, let's say, pushed 1074 00:37:41,114 --> 00:37:43,806 off track, they will stay on that off 1075 00:37:43,828 --> 00:37:45,906 track path for a longer period of time. And 1076 00:37:45,928 --> 00:37:47,506 then there will be feedback loops with, 1077 00:37:47,528 --> 00:37:49,778 let's say, the liver and other organs to 1078 00:37:49,944 --> 00:37:51,830 just try to adapt 1079 00:37:51,830 --> 00:37:53,890 with the sort of external trigger. 1080 00:37:53,890 --> 00:37:55,998 So that was sort of part 1081 00:37:56,024 --> 00:37:58,566 of our sort of understanding of 1082 00:37:58,748 --> 00:38:00,966 the use of drugs. But again, it 1083 00:38:00,988 --> 00:38:02,886 showed that taking medication has 1084 00:38:02,908 --> 00:38:04,854 a quite substantial effect on 1085 00:38:04,892 --> 00:38:06,854 your proteome. And we 1086 00:38:06,892 --> 00:38:08,646 found it fascinating that it actually seemed 1087 00:38:08,678 --> 00:38:10,902 to be consistent 1088 00:38:10,902 --> 00:38:12,874 over many years, a 1089 00:38:12,912 --> 00:38:14,682 picture of real time biology, the 1090 00:38:14,736 --> 00:38:16,170 proteome, right. 1091 00:38:16,320 --> 00:38:18,602 And, what we 1092 00:38:18,656 --> 00:38:20,970 know about effects of certain 1093 00:38:21,040 --> 00:38:23,406 treatments, what we know about effects of 1094 00:38:23,428 --> 00:38:25,246 certain drugs, we're just scratching the 1095 00:38:25,268 --> 00:38:27,914 surface. Right. A number of our pharma 1096 00:38:27,962 --> 00:38:29,934 partners and customers of Olink 1097 00:38:29,972 --> 00:38:31,934 are finding out so much with just 1098 00:38:31,972 --> 00:38:33,754 a limited set of proteins. 1099 00:38:33,882 --> 00:38:35,778 They're not looking at the proteome, they 1100 00:38:35,784 --> 00:38:37,362 might be looking at a panel 1101 00:38:37,416 --> 00:38:39,986 of 50 or 90 [proteins], or what 1102 00:38:40,008 --> 00:38:42,386 have you. But there is just so much to 1103 00:38:42,408 --> 00:38:44,390 learn about the biology. 1104 00:38:44,390 --> 00:38:46,598 Sarantis, you started to ask, I think you 1105 00:38:46,604 --> 00:38:48,646 were down a path of a couple of questions. I 1106 00:38:48,668 --> 00:38:50,970 was curious. 1107 00:38:50,970 --> 00:38:52,694 I want 1108 00:38:52,732 --> 00:38:54,758 to ask just a more 1109 00:38:54,844 --> 00:38:56,598 philosophical question about 1110 00:38:56,764 --> 00:38:58,418 if you have some, let's 1111 00:38:58,434 --> 00:39:00,562 say, mother, 1112 00:39:00,562 --> 00:39:02,906 sisters, some relatives, or if 1113 00:39:02,928 --> 00:39:04,778 you have some twins that you can follow. 1114 00:39:04,864 --> 00:39:06,634 And you can see the change comes from 1115 00:39:06,672 --> 00:39:08,346 genetic background, or comes from the 1116 00:39:08,368 --> 00:39:10,490 proteomics background, or combination, or 1117 00:39:10,560 --> 00:39:12,926 neither. Do you have any experience on 1118 00:39:12,948 --> 00:39:14,874 that? Have you seen some patterns 1119 00:39:14,922 --> 00:39:16,770 around [that]? 1120 00:39:16,770 --> 00:39:18,622 I don't think we 1121 00:39:18,676 --> 00:39:20,860 necessarily looked into this. 1122 00:39:20,860 --> 00:39:22,786 But I've been working with 1123 00:39:22,808 --> 00:39:24,846 another twin cohort from Sweden called Twin 1124 00:39:24,878 --> 00:39:26,950 Gene, which the name says 1125 00:39:26,950 --> 00:39:28,930 has a quite clear 1126 00:39:29,000 --> 00:39:31,200 focus on these aspects. 1127 00:39:31,200 --> 00:39:33,620 And 1128 00:39:33,620 --> 00:39:35,830 no, 1129 00:39:35,830 --> 00:39:37,958 not that I think in particular, but of 1130 00:39:37,964 --> 00:39:39,974 course, it's, again, what we 1131 00:39:40,012 --> 00:39:42,582 pass on to our children is something 1132 00:39:42,636 --> 00:39:44,982 that will be, in the future, 1133 00:39:45,036 --> 00:39:47,540 helpful for them to know. 1134 00:39:47,540 --> 00:39:49,702 And maybe 1135 00:39:49,836 --> 00:39:51,514 they will change their lives when they know. 1136 00:39:51,552 --> 00:39:53,642 Okay. I'm at a higher risk of a certain 1137 00:39:53,696 --> 00:39:55,942 disease because both my parents 1138 00:39:56,086 --> 00:39:58,314 passed away. I guess you see a lot of these 1139 00:39:58,352 --> 00:40:00,926 breast cancer studies and effects in 1140 00:40:00,948 --> 00:40:02,900 Iceland, I think, right. 1141 00:40:02,900 --> 00:40:04,760 But 1142 00:40:04,760 --> 00:40:06,906 not in these studies. I cannot 1143 00:40:06,938 --> 00:40:08,734 recall that we actually specifically looked 1144 00:40:08,772 --> 00:40:10,814 into this. Yeah. What was I think really 1145 00:40:10,852 --> 00:40:12,762 interesting about this particular paper 1146 00:40:12,836 --> 00:40:14,642 on breast cancer is that you looked 1147 00:40:14,696 --> 00:40:16,766 into so many different kinds 1148 00:40:16,798 --> 00:40:18,750 of connections 1149 00:40:18,910 --> 00:40:20,798 in terms of inherited risk 1150 00:40:20,894 --> 00:40:22,846 as well because the title 1151 00:40:22,878 --> 00:40:24,818 is paper, "Evaluation of 1152 00:40:24,824 --> 00:40:26,686 Circulating Plasma Proteins in Breast 1153 00:40:26,718 --> 00:40:28,738 Cancer and Mendelian Randomization 1154 00:40:28,834 --> 00:40:30,710 Analysis," you're actually 1155 00:40:30,780 --> 00:40:32,866 looking at, then, the entire genetic 1156 00:40:32,898 --> 00:40:34,514 backgrounds of unrelated 1157 00:40:34,562 --> 00:40:36,854 individuals and just saying what 1158 00:40:36,892 --> 00:40:38,778 is elevating that particular 1159 00:40:38,864 --> 00:40:40,874 risk. And understand, these five 1160 00:40:40,912 --> 00:40:42,694 proteins that were differentially 1161 00:40:42,742 --> 00:40:44,990 regulated were basically 1162 00:40:44,990 --> 00:40:46,966 lifetime 1163 00:40:46,966 --> 00:40:48,922 exposures. That a person 1164 00:40:48,976 --> 00:40:50,470 was exposed to a high level of 1165 00:40:50,480 --> 00:40:52,878 protein throughout their whole life. And I 1166 00:40:52,884 --> 00:40:54,870 think that's what makes this really 1167 00:40:54,870 --> 00:40:56,590 fascinating, right? The 1168 00:40:56,660 --> 00:40:58,846 proteomics being informed by the 1169 00:40:58,868 --> 00:41:00,926 genetics controlling the levels of 1170 00:41:00,948 --> 00:41:02,814 protein, and then saying these 1171 00:41:02,852 --> 00:41:04,766 five proteins actually become drug 1172 00:41:04,798 --> 00:41:06,914 targets, which I thought was 1173 00:41:06,952 --> 00:41:08,782 just a fascinating realm. 1174 00:41:08,820 --> 00:41:10,946 Before we wrap up, Jochen, would you 1175 00:41:10,968 --> 00:41:12,542 like to make any final comments? 1176 00:41:12,686 --> 00:41:14,980 Either, I don't know, about 1177 00:41:14,980 --> 00:41:16,838 where we are, 1178 00:41:16,924 --> 00:41:18,966 where we're going, working with 1179 00:41:18,988 --> 00:41:20,886 Olink? Oh, I understand, right? We 1180 00:41:20,908 --> 00:41:22,822 didn't even talk about a very 1181 00:41:22,876 --> 00:41:24,250 famous 1182 00:41:24,250 --> 00:41:26,450 postdoc, famous at Olink, 1183 00:41:26,530 --> 00:41:28,438 Philippa [Pettingill] came out of your lab. I don't know 1184 00:41:28,444 --> 00:41:30,234 if you want to talk about what it was like 1185 00:41:30,272 --> 00:41:32,918 working with her. She has helped 1186 00:41:32,934 --> 00:41:34,534 me, Cindy, with her title. She's 1187 00:41:34,582 --> 00:41:36,726 Director [of Application Sciences]. She's a superstar. 1188 00:41:36,838 --> 00:41:38,314 She runs the field application 1189 00:41:38,432 --> 00:41:40,622 scientist team within 1190 00:41:40,756 --> 00:41:42,538 the European 1191 00:41:42,714 --> 00:41:44,350 region, and she 1192 00:41:44,420 --> 00:41:46,826 is absolutely magnificent. 1193 00:41:46,858 --> 00:41:48,960 She's also helped lead 1194 00:41:48,960 --> 00:41:50,440 our discussions around 1195 00:41:50,440 --> 00:41:52,818 statistical analyses in the 1196 00:41:52,824 --> 00:41:54,926 UK Biobank Project. She's 1197 00:41:54,958 --> 00:41:56,770 just such a magical 1198 00:41:56,770 --> 00:41:58,754 human being to 1199 00:41:58,792 --> 00:42:00,898 have at Olink. We're so lucky to 1200 00:42:00,904 --> 00:42:02,834 have her. And she is a product 1201 00:42:02,952 --> 00:42:04,710 launched 1202 00:42:04,710 --> 00:42:06,934 out of your lab. At 1203 00:42:06,972 --> 00:42:08,810 some point, 1204 00:42:08,810 --> 00:42:10,582 you had an impact on 1205 00:42:10,716 --> 00:42:12,918 her trajectory. So, yes, 1206 00:42:13,004 --> 00:42:15,974 please, anything you have to say about her 1207 00:42:16,012 --> 00:42:18,898 would be greatly appreciated. We had 1208 00:42:18,924 --> 00:42:20,938 hoped we'd be able to have her on, but 1209 00:42:20,944 --> 00:42:22,506 we weren't able to get her 1210 00:42:22,528 --> 00:42:24,940 into the timing that we had 1211 00:42:25,000 --> 00:42:27,866 going. No, I 1212 00:42:27,888 --> 00:42:29,610 mean, all the success that 1213 00:42:29,680 --> 00:42:31,606 she has now is because 1214 00:42:31,648 --> 00:42:33,646 of her 1215 00:42:33,668 --> 00:42:35,262 engagement, her knowledge, and her 1216 00:42:35,316 --> 00:42:37,898 curiosity. But, yeah, it was fantastic 1217 00:42:37,914 --> 00:42:39,822 to work with her. She was with me about 1218 00:42:39,876 --> 00:42:41,854 one and a half, two years. It 1219 00:42:41,892 --> 00:42:43,430 was 1220 00:42:43,430 --> 00:42:45,990 inspirational and fun 1221 00:42:45,990 --> 00:42:47,762 from the first to the last day. 1222 00:42:47,816 --> 00:42:49,750 And I think 1223 00:42:49,750 --> 00:42:51,890 to see someone leaving the lab 1224 00:42:51,960 --> 00:42:53,586 and making such a 1225 00:42:53,608 --> 00:42:55,750 wonderful career 1226 00:42:55,750 --> 00:42:57,986 is fantastic. I guess if my 1227 00:42:58,008 --> 00:43:00,294 contribution is that I showed her all these 1228 00:43:00,332 --> 00:43:01,990 different tools that we had in lab, 1229 00:43:02,060 --> 00:43:04,038 including Olink and others, and we talked a 1230 00:43:04,044 --> 00:43:05,526 lot about the different assays, the 1231 00:43:05,548 --> 00:43:07,974 different concepts. So if that has 1232 00:43:08,012 --> 00:43:10,326 helped her in achieving these 1233 00:43:10,348 --> 00:43:12,806 fantastic things that she's doing with 1234 00:43:12,828 --> 00:43:14,666 you, it makes me proud and 1235 00:43:14,688 --> 00:43:16,922 happy. I think she deserves it. 1236 00:43:16,976 --> 00:43:18,906 And I wish, of course, her 1237 00:43:18,928 --> 00:43:20,800 all the success, and 1238 00:43:20,800 --> 00:43:22,926 anytime we see her, we see 1239 00:43:22,948 --> 00:43:24,586 each other on the media calls, 1240 00:43:24,698 --> 00:43:26,860 it's like old friends. 1241 00:43:30,860 --> 00:43:32,590 I think she came out 1242 00:43:32,660 --> 00:43:34,746 absolutely a leader, and I think 1243 00:43:34,868 --> 00:43:36,674 she has such great things to say 1244 00:43:36,712 --> 00:43:38,834 about the time that she spent in your 1245 00:43:38,872 --> 00:43:40,894 lab. And I think that's 1246 00:43:40,942 --> 00:43:42,802 pretty sweet. Thank you. 1247 00:43:42,856 --> 00:43:44,994 That's great to hear. All right, 1248 00:43:45,032 --> 00:43:46,754 well, thank you very much for joining us 1249 00:43:46,792 --> 00:43:48,614 today, Jochen. We've really enjoyed the 1250 00:43:48,652 --> 00:43:50,966 conversation. Thank you for having me. 1251 00:43:50,988 --> 00:43:52,300 It was fantastic. And 1252 00:43:52,300 --> 00:43:54,834 continue with this great podcast. 1253 00:43:54,962 --> 00:43:56,614 It's really a treasure. Thanks a lot 1254 00:43:56,652 --> 00:43:58,502 for setting this up and running 1255 00:43:58,556 --> 00:44:00,454 it. You're so 1256 00:44:00,492 --> 00:44:02,774 kind. Thank you. 1257 00:44:02,972 --> 00:44:04,774 Okay, well, I think 1258 00:44:04,812 --> 00:44:06,760 that's it. Thank you. 1259 00:44:08,760 --> 00:44:10,950 1260 00:44:10,950 --> 00:44:12,974 Thank you for 1261 00:44:13,012 --> 00:44:14,794 listening to the Proteomics in Proximity 1262 00:44:14,842 --> 00:44:16,746 podcast brought to you by Olink 1263 00:44:16,778 --> 00:44:18,826 Proteomics. To contact the hosts 1264 00:44:18,858 --> 00:44:20,714 or for further information, simply 1265 00:44:20,762 --> 00:44:22,860 email info@olink.com.