Monday, December 30, 2013

The Gene Factory (New Yorker)


A Chinese firm’s bid to crack hunger, illness, evolution—and the genetics of human intelligence.


A well-balanced article about BGI in the New Yorker.

I am misquoted about our current ability to predict height from genomic information, despite an hour on the phone with the Harvard-educated fact checker. There is also some confusion in that paragraph concerning the correlation between height and IQ and its relevance to the broader discussion  :-(

Chris Chang has the wise, last words in the article :-)  Click for larger version.


LondonYoung said...

"people will look back and wonder what all the fuss was about" - indeed.
So, Jupiter is almost at opposition right now, and last night I had the telescope out (my devious plot to interest the kids in science). It is very easy to see the four Galilean moons and since Io has an orbital period of under two days it is easy to see that it moves (and along the ecliptic 0 of course). So, since about 1610 is has been pretty clear that Jupiter was the center of the this mini-solar system, yet it took more than 100 years for the Copernican view of the solar system to be accepted. Sigh. So what was all the fuss about?

(In fact, if you have a nice 8 inch telescope, you can see Io's shadow when it crosses in front of Jupiter so that it is painfully clear what is going on. But I am not sure when the shadows were first observed.)

steve hsu said...

While I believe that people will look back and wonder what all the fuss was about, I don't believe the consequences of the new technologies will be anything short of transformative in the medium to long run. I'm very confident that there are many, perhaps ~ 10-30, SDs up for grab here.

twoL said...

Do you think those gains are more likely to occur through embryo selection or a Crispr like technique?

5371 said...

To be so excited about this stuff, you must surely assign an extremely low probability to a silicon-based singularity. But until recently you knew much more about IT than microbiology, isn't that so?

jeffhsu3 said...

Here is a thought, a silicon-based singularity will be quite expensive and if you plan on beating organizations such as Google or Amazon to it, you are probably going to have to spend billions of dollars. These tech companies have already dozens of billions of dollars invested in large data centers and compute clusters. I can't begin to estimate when a silicon based singularity would occur, but if you were to hedge that it doesn't within the next 30, creating 30 SD humans would be a good bet a securing the resources to do so.

Jorge_Videla said...

absurd. all the risk factors for cvd and atherosclerosis related diseases are as heritable as iq, but one very simple intervention reduces them all to 0!,,

get over yourself paul. ayn rand is for adolescents, like "the hobbit".

and btw, whether the earth goes round the sun or the sun round the earth is purely a practical matter. it's just more easier and therefore more useful to think of it in the former way than the latter.

Jorge_Videla said...

when did you start with the mushrooms steve? was it five years ago when you claimed that today the genes for intelligence would be revealed.

"if you want a picture of the future winston, imagine a boot stamping on a human face, forever."

the hereditists have become the new lysenkos.

jeffhsu3 said...

I'll point out that a lot of other data has contradicted caloric restriction (CR) (something I practice sporadically but not by choice) in the long-term. See:
CR also isn't something simple, and if it were, new year's diet resolutions wouldn't be an annual thing.

Also note that statins offer a much stronger response than what is presented in that study. Statins are the partial result of genetic studies (along with many other biomedical disciplines), so the notion that studying genetics (as you seem to imply but can't articulate) is a strawman.

steve hsu said...

CRISPR is looking very interesting.

steve hsu said...

In my opinion, no singularity for 50+ years. Most people who use the term singularity are incredibly over-optimistic about it.

disqus_LUJWSJEndf said...

Does CRISPR involve discarding (to be more politically correct) embryos? The wikipedia article seems to say that it doesn't, but being wikipedia and not being a biologist capable of understanding their jargon I may be wrong.

5371 said...

High intelligence is neither necessary nor sufficient for the amassing of wealth, and in fact doesn't help very much. If that is true for individuals, how much more for societies.

steve hsu said...

CRISPR allows direct gene editing. Read the links on this blog, not Wikipedia.

Max B said...

Thats what I thought. If you were as excited about AI as,say, Kurzweil. There is no way you would continue looking in the direction of wetware (which is what genetic engineering ultimate goal is). I think this is rather antropocentric point of view though. Once AI passes the treshold of AGI (artificial general intelligence) - it is over. It doesnt need years to learn or improve itself. It will use existing resources and all it might take a few hours for it to far surpass the entire historical progress of human civilization

JorgeVidela said...

not the case. they were indistinguishable from controls prior to commencing cr. click on the last link and read the intro.

JorgeVidela said...

"so the notion that studying genetics (as you seem to imply but can't articulate) is a strawman."

sorry, i don't understand that clause.

the effect of genes is with rare exceptions mediated by the environment to such an extent, as the above study demonstrates, that whatever the results they are meaningless.

steve thinks he can breed a pole vaulter who can vault twice as high as bubka and a sprinter twice as fast as bolt. this is all rot to anyone who isn't on mushrooms.

DK said...

Health-obsessed individuals appear healthier than controls. Shocking!

DK said...

Be specific. The way you prognosticate is no different from a fortune teller.

Here is a baseline for you: targeted immunotherapy and gene therapy have been hyped for at least 25-30 years. Today, the former has finally succeeded *in few isolated cases* and the latter is still very much experimental in all cases. Both of these are orders of magnitude less complex than the new technologies you are thinking of.

DK said...

Yes, if it would ever be applied to editing genes in embryos, it would necessarily involve discarding embryos on a massive scale. The only alternative to that (bar the fantasies that Steve and Kurzweil bear) is coupling genome editing with human stem-cells based cloning.

jeffhsu3 said...

It's true that all the heritability is in the g-factor, but where I'm coming from (I'm guessing I am more of a g-man than Steve is), shooting up g would be the goal (or other sub components), if the multitude of testing could be done.

Also the genetic architecture of pole vaulting and sprinting is probably a lot different than cognitive capabilities. I highly doubt they are comparable because no one knows yet for either traits. I can't even find the variance for traits like pole vaulting in the general population. 10-30 SD seem like much but we've had many such gains throughout history.

jeffhsu3 said...

The correlation between intelligence as measured by IQ and income is roughly 0.4, not high, but certainly actionable given technology or time. Obviously a lot of studies also point to conscientiousness and other personality factors having a similar effect, and those have genetic contributions as well. So the point is there are changes you can make if you want *slightly* increase your chances of achieving the singularity over others (not only wealth, but having good programmers would certainly be a benefit).

Of course this argument is likely academic. The singularity event will be a winner take all so the window where genetic engineering has any impact likely very small but may be worth it depending on your priors.

Magnusmaster said...

But with cloning wouldn't you still have to discard the clones?

JorgeVidela said...



JorgeVidela said...



≈1 yr CR

BMI, kg/m2

24.5 ± 2.6
20.9 ± 2.4
19.5 ± 2.1

Tchol, mg/dl
194 ± 45
161 ± 31
157 ± 38

LDL-C, mg/dl
122 ± 36
89 ± 24
86 ± 17

HDL-C, mg/dl
43 ± 8
58 ± 13
65 ± 24

Tchol/HDL-C ratio
4.1 ± 1
2.8 ± 0.5
2.5 ± 0.4

TG, mg/dl
149 ± 87
72 ± 35
54 ± 15

Systolic BP, mmHg
132 ± 15
112 ± 12
97 ± 8

Diastolic BP, mmHg
80 ± 11
69 ± 7
59 ± 5

Bobdisqus said...

Simple? If you think so try to get your family to do it with you for a year. Then come tell us about it.

LondonYoung said...

I did read the intro, but that's not a controlled experiment, right? There are quite possibly third factors which both improve the odds of adopting CR and improve the health outcome. Factoring out initial cholesterol and BP levels doesn't even scratch the surface of possible third factors. If you selected a random group and flipped a coin to see who would be CR and who wouldn't then you are exposed only to statistical noise. But if you let people self select how do you establish cause and effect?

DK said...

Yes, you would. But potentially, less of them. Under this scenario, most of the editing/screening is done on unfertilized cells.

David Coughlin said...

We got Stella Snap Circuits [ ] . Should be good for a couple of years of visiting and revisiting.

JorgeVidela said...

if you'd clicked the last link you'd see all the stats for the cr group prior to commencing cr. fontana has described his subjects as like another species. another paper of his shows that these numbers CANNOT be got through excercise, even though heavy exercisers tend to have even lower body fat %. it could be that for all 18 there was some third factor which coincided with commencing cr. it could also be that sasquatch and the yeti and professional wrestling are real. it could be.

the point is that even for physical and completely objective traits (risk factors for atherosclerosis related disease in this case)which are in many studies as heritable as IQ and a fortiori as heritable as any other psychological traits, a very simple (but difficult) environmental change makes all who commence it at the "ceiling" of low risk, so to speak.

hereditists are the new lysenkos. they are to neo-liberalism what lysenko was to stalin. both are wrong. thankfully the hereditism has begun to ebb.

jeffhsu3 said...

risk factors for atherosclerosis as heritable as IQ.

As a cardiovascular researcher I wish I could say that is true. It's about half. Also I love how one paper in PNAS justifies your stupidity.

LondonYoung said...

I agree with the spirit of what you are saying/concluding, but I think it is much vastly less likely that bigfoot exists than that a third factor exists which both helps people adopt and stay on CR and, independently of that, also indicates better health outcome. I think there is a great hypothesis here, but it remains untested via the scientific method - as least as physicists are inclined to interpret it.

But to be clear, I have no doubt that choosing to relax for our distant ancestors often meant taking it easy and being a bit hungry rather than sitting in front of the TV and pigging out. Of course, heart disease wasn't a big concern for those distant ancestors anyway ....

JorgeVidela said...

"also indicates better health outcome"

the "health outcome" was just the (very dramatic) change in risk factors. although individual studies are meaningless, together they all point in one direction regarding diet and health, physical activity and health.

Parameter Pre-CR ?1 yr CR Present
BMI, kg/m2 24.5 ± 2.6 20.9 ± 2.4 19.5 ± 2.1
Tchol, mg/dl 194 ± 45 161 ± 31 157 ± 38
LDL-C, mg/dl 122 ± 36 89 ± 24 86 ± 17
HDL-C, mg/dl 43 ± 8 58 ± 13 65 ± 24
Tchol/HDL-C ratio 4.1 ± 1 2.8 ± 0.5 2.5 ± 0.4
TG, mg/dl 149 ± 87 72 ± 35 54 ± 15
Systolic BP, mmHg 132 ± 15 112 ± 12 97 ± 8
Diastolic BP, mmHg 80 ± 11 69 ± 7 59 ± 5

BobSykes said...

The image on the cover of the New Yorker is clearly a Communist symbol. Are we supposed to understand that the revolution in genetic engineering will result in the ultimate Communist victory?

JorgeVidela said...

it wasn't much of a concern to your recent ancestors either. from its beginning to 1800 life expectancy at birth in the civilized world never exceeded that of savages. in fact, it was less. the archeological record shows the transition to agriculture resulted in higher mortality, shorter stature, rotten teeth, etc.

JorgeVidela said...

i said it was simple not easy. sometimes simple is used to mean easy. look up "polysemy" bob.

but it's not only simple, it's easy. just eat like a chimp or a caveman. those following a paleo diet spontaneously reduced their calorie intake by 30% in one study. the restriction is based on your ad lib intake. some need more than others to maintain the same weight. if you maintain a bmi of 20 with little exercise you're doing it.

JorgeVidela said...

in fact, being a herr doktor of physics, can the orbital acceleration of the earth be measured directly on earth? that is, without reference to a model, does earth's orbital acceleration move anything on earth?

if so then there would be a strange force left unexplained by the geocentric model, not just the loopdeloops of the planets.

"a man who is falling does not feel his own weight" suggests the answer is actually no.

JorgeVidela said...

one paper? which one. cr has extended life expectancy at least, and in some cases lifespan for spiders, fruit flies, mice, rats, and the wisc studies shows monkeys.

the heritability for bp, bmi, hdl and ldl of various sorts, hs crp, etc. all have heritabilities > .6 in some studies. that studies disagree just shows that heritability itself is meaningless.

you're the stupid one jeff. you're wasting your time on cvd research. cvd is a disease of civilization. why don't you research the genetics of susceptibility to tobacco related cancers. i have zero doubt that some are more susceptible than others.

JorgeVidela said...

"Also the genetic architecture of pole vaulting..."

the point jeff, was that a human who could vault twice as high as bubka or run twice as fast as bolt would have to be NOT human. a person with an iq of 250 = 10 sds would look like this: NOT like this:

LondonYoung said...

it might come up in developing a model for the ocean tides

Blog Archive