![[personal profile]](http://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
(no subject)
Jan. 21st, 2014 06:36 amReturning from California where I was at the 10th International Imaging Genetics Conference, this past weekend. I had to split before the end of it this morning—missed one of the more interesting talks, I'm betting—but what I saw of it was excellent as usual.
I'm so happy that this conference has kept going. I remember in 2003 when the boss said we should hold an imaging genetics conference, and I said you can't just start a conference! And he said, in effect, watch me. :) I busted my butt for that first conference, and the next two less so, and then it was firmly in a co-worker's hands—and she's still working it every year and does a fantastic job. The boss manages to line up a bunch of excellent speakers every year—sometimes he gets a little repetitive, having people back to give talks again, but there's always somebody new and a new theme for a couple of the talks.
And this one is the TENTH. ::chuckle:: A field has grown up! It's totally cool to have been in on it so early, and to see where it is going and why and how. It's one of the more promising research domains, in my not so humble opinion—it's where people have to collaborate, if anything is going to get done at all. Especially in this era of NIH not funding massive projects any more, people have to share their smaller datasets if we are ever going to find anything. The Psychiatric Genetics Consortium (of which I am not a part, sigh and foo) is doing genome-wide scan meta-analyses of 85,000 cases and controls from around the world, and it's just so exciting to see! Behind them is the ENIGMA consortium (of which I am a part, yay and wahoo!) doing imaging and genetics meta-analyses on 10,000 subjects, or smaller sub-samples of that for various questions.
On the one hand, you'd think that with that number of subjects, we'd be able to nail genes' and brains' relationships, right? On the other hand, the problem with consortia and legacy data is that you end up only being able to analyze the lowest common denominator, the variables that everyone collected. Everyone usually can tell you the age and gender of their subjects, and whether they had schizophrenia or not; but did they collect info about medication levels? Age of disease onset? Whether they were also type II diabetics? What their clinical symptoms were like? Everyone has something but nobody has everything, and a lot of people don't have much, which quickly cuts down your sample sizes.
But nonetheless, we persevere. I said to Theva, who I'm trying to write up the ENIGMA stuff with, that the large consortia doing meta-analyses form the basement; they aren't the glamorous first floor that the public should see, but they pull together all the heterogeneous results that you get from running smaller samples around the world, and can at least make statements about the bulk of the evidence one way or the other. Then you can do the more focused, elegant experiments that target specific hypotheses.
But anyway, personal history: I flew out Sunday morning, got to California around lunch time. Was going to spend the afternoon working on lecture notes, but got pulled into the paper with Theva and after swapping a bunch of emails finally he sent me an email saying how about I pick you up and we can work on it in the office. :) So we did that for a few hours, and then I was back to the hotel, to have dinner with Pash and Mysh, and Pa and Ny as well! They took me to a dosa restaurant that the hubby and I used to love—it's still open, amazingly enough, though it's been handed down to the next generation in the family and they are pulling back a little from the dosa emphasis. Still tasty, though! And it was good to catch up with everybody. I was back at the hotel and sound asleep before 9 pm.
Slept like a log but was up by 5ish, which let me get caught up on some things and go for a short run in the hotel gym before heading to the conference at 7:30. Bumped into Peko in the elevator on the way down, which was nice—I haven't known him as long as I have Dake and Jebo, but I always enjoy my conversations with him; he's a serious scientist who pokes holes in my arguments with some regularity, but he's also really funny. I got to see Jebo and some others over breakfast whom I haven't seen in years, and that was great.
The conference went on all day yesterday and into the evening; I had lunch with Eswa, who's finishing her PhD this spring and angling to convince her government to give her funds to do a post-doc with me, which would be awesome. I hope we can do that!! She'd enjoy Atlanta, and she's great to work with.
I was juggling making lecture slides for Wednesday while listening to the talks and taking notes, which was a little distracting. But it worked out, and I actually asked some questions at the various panel discussions, which if it wasn't the first time may only have been the second (so yay me!). The poster session at the end of the day was a little exhausting, so I led Jebo and Peko and Cawi to the 7 pm shuttle so we could get back to the hotel and have dinner. It was a quiet but late evening—no raucous pool-playing or darts, just dinner in the hotel restaurant with about six of us. Jebo and I hung out for an extra round of drinks and chatting after everyone else called it a night, and wrapped it up around 11.
So this morning I got to talk to the old boss over breakfast, which was an unexpected treat. :) Then I heard the first two talks—unfortunately the first guy was heavily accented and I couldn't make out what he was talking about, but the second guy was excellent. He could give the history of the PGC growth and successes, and from that back into the more detailed, focussed experiments that their findings have spawned, and some of the complexities that's led to in the understanding about some of these genes' effects. Really good stuff. I caught a ride with Jebo to the airport after that, though. And here we are. :)
I'm so happy that this conference has kept going. I remember in 2003 when the boss said we should hold an imaging genetics conference, and I said you can't just start a conference! And he said, in effect, watch me. :) I busted my butt for that first conference, and the next two less so, and then it was firmly in a co-worker's hands—and she's still working it every year and does a fantastic job. The boss manages to line up a bunch of excellent speakers every year—sometimes he gets a little repetitive, having people back to give talks again, but there's always somebody new and a new theme for a couple of the talks.
And this one is the TENTH. ::chuckle:: A field has grown up! It's totally cool to have been in on it so early, and to see where it is going and why and how. It's one of the more promising research domains, in my not so humble opinion—it's where people have to collaborate, if anything is going to get done at all. Especially in this era of NIH not funding massive projects any more, people have to share their smaller datasets if we are ever going to find anything. The Psychiatric Genetics Consortium (of which I am not a part, sigh and foo) is doing genome-wide scan meta-analyses of 85,000 cases and controls from around the world, and it's just so exciting to see! Behind them is the ENIGMA consortium (of which I am a part, yay and wahoo!) doing imaging and genetics meta-analyses on 10,000 subjects, or smaller sub-samples of that for various questions.
On the one hand, you'd think that with that number of subjects, we'd be able to nail genes' and brains' relationships, right? On the other hand, the problem with consortia and legacy data is that you end up only being able to analyze the lowest common denominator, the variables that everyone collected. Everyone usually can tell you the age and gender of their subjects, and whether they had schizophrenia or not; but did they collect info about medication levels? Age of disease onset? Whether they were also type II diabetics? What their clinical symptoms were like? Everyone has something but nobody has everything, and a lot of people don't have much, which quickly cuts down your sample sizes.
But nonetheless, we persevere. I said to Theva, who I'm trying to write up the ENIGMA stuff with, that the large consortia doing meta-analyses form the basement; they aren't the glamorous first floor that the public should see, but they pull together all the heterogeneous results that you get from running smaller samples around the world, and can at least make statements about the bulk of the evidence one way or the other. Then you can do the more focused, elegant experiments that target specific hypotheses.
But anyway, personal history: I flew out Sunday morning, got to California around lunch time. Was going to spend the afternoon working on lecture notes, but got pulled into the paper with Theva and after swapping a bunch of emails finally he sent me an email saying how about I pick you up and we can work on it in the office. :) So we did that for a few hours, and then I was back to the hotel, to have dinner with Pash and Mysh, and Pa and Ny as well! They took me to a dosa restaurant that the hubby and I used to love—it's still open, amazingly enough, though it's been handed down to the next generation in the family and they are pulling back a little from the dosa emphasis. Still tasty, though! And it was good to catch up with everybody. I was back at the hotel and sound asleep before 9 pm.
Slept like a log but was up by 5ish, which let me get caught up on some things and go for a short run in the hotel gym before heading to the conference at 7:30. Bumped into Peko in the elevator on the way down, which was nice—I haven't known him as long as I have Dake and Jebo, but I always enjoy my conversations with him; he's a serious scientist who pokes holes in my arguments with some regularity, but he's also really funny. I got to see Jebo and some others over breakfast whom I haven't seen in years, and that was great.
The conference went on all day yesterday and into the evening; I had lunch with Eswa, who's finishing her PhD this spring and angling to convince her government to give her funds to do a post-doc with me, which would be awesome. I hope we can do that!! She'd enjoy Atlanta, and she's great to work with.
I was juggling making lecture slides for Wednesday while listening to the talks and taking notes, which was a little distracting. But it worked out, and I actually asked some questions at the various panel discussions, which if it wasn't the first time may only have been the second (so yay me!). The poster session at the end of the day was a little exhausting, so I led Jebo and Peko and Cawi to the 7 pm shuttle so we could get back to the hotel and have dinner. It was a quiet but late evening—no raucous pool-playing or darts, just dinner in the hotel restaurant with about six of us. Jebo and I hung out for an extra round of drinks and chatting after everyone else called it a night, and wrapped it up around 11.
So this morning I got to talk to the old boss over breakfast, which was an unexpected treat. :) Then I heard the first two talks—unfortunately the first guy was heavily accented and I couldn't make out what he was talking about, but the second guy was excellent. He could give the history of the PGC growth and successes, and from that back into the more detailed, focussed experiments that their findings have spawned, and some of the complexities that's led to in the understanding about some of these genes' effects. Really good stuff. I caught a ride with Jebo to the airport after that, though. And here we are. :)
