Skip to content

Is there a geneticist in the house?

January 10, 2010

I feel like I have a tenuous grasp on what heritability means and why it is *not* the same thing as saying a trait is genetic.

And yet, reading about this new study on the genetic causes of obesity, I’m confused again. I only have access to the abstract (booooo) but the conclusion, which Marion Nestle talks about in this post, is that the different gene variants associated with increased bmi only account for about 1% of differences in weight.

So how does it work that a trait can have very high heritability, but that same trait have a fairly low correlation with gene variants? Do I just need to take a class? Help.

ETA: This post dealing with the heritability of IQ is a great, though very very long, look at the concept of heritability. Well worth the read if you find these things as interesting as I do.

5 Comments leave one →
  1. January 11, 2010 7:34 am

    I’m not a geneticist but I’ll try to help.

    I think you’ve come across one of the stumbling point when dealing with research articles. That is, science only asks one question at a time.

    So. The first part is to sort out WHAT the study your reading about is asking. It took me a bit to sort out the question being asked but I think I have it figured out. And the question being asked is very specific.

    At some point in the past, they found 12 genetic loci (aka gene segments) associated with BMI. This study is tested those 12 genetic loci and determined those SPECIFIC gene variants are only 1% associated with BMI.

    The problem is that Marion Nestle is then using this study to support her argument that fat is about behavior, not genetics. When, in reality, it’s just ONE piece in a larger puzzle.

    As for the second part, the article you linked to about heritability seemed more confusing than helpful.

    I think the guy was just arguing semantics. That the definition of heritability isn’t about an individuals genetic makeup, it’s really a calculation of genetic variation within a population.

    Regardless, “80% heritable” is still referencing environment vs. genetics.

    I understand that the guy is upset that people are using the term wrong, but I don’t think he succeeds in making the matter any less confusing.

    • January 11, 2010 1:01 pm

      Oh, I really liked his explanation. It’s one of the first I’ve come across that made sense to me.

      My question was based on an assumption that, because this was written quickly, wasn’t explicitly stated. I read a paper many months ago (which I can’t find because backflip is down) that stated that scientists had found all the genetic loci with the largest impact on BMI and that while they might find more down the road, none would be correlated as strongly. Now, assuming this is true (big assumption) my question was more: how would it work that genes overall had a low correlation with BMI and yet heritability of BMI was fairly high?

      Unfortunately, I’m neither well-versed in genetics or statistics beyond intro level classes so I’ve hit a wall. It may be that this paper only focused on a few of the many genes found – although any decent paper should definitely have used the genes with the strongest correlation to BMI. Or it could be that mathematically, the low genetic contribution/high heritability thing works. Unfortunately, I don’t have access to the paper so I can’t do much other than wonder.

  2. hsofia permalink
    January 19, 2010 7:21 pm

    @elizabethturnquist – I would think genetics also play a role in behaviors. I suspect there are many dozens (maybe hundreds) of genes that affect how fat/thin we are. That’s why, ultimately, I think it’s a little pointless to worry about! =)

  3. November 23, 2010 8:25 pm

    Well, here’s what I can make of it (disclaimer: I’m not reading a long paper on heritability of IQ right now, but I took a quick look at the discover magazine link and it is in line with my knowledge of the subject):
    -The researchers studied previously-identified genes associated with BMI and found only a 1% influence on the variation in BMI
    -These people were all in presumably a similar environment, WRT effects on obesity
    -Twin studies indicate obesity is 70-80% heritable (WITHIN a given environment), and for heritability they are NOT talking about effects of parental behavior influence, because these are twins raised apart–in different homes, by different parents. (I believe these studies do use the term “heritable” in describing their results.) However, similarity between twins could also be affected by things like womb environment and genetic imprinting/epigenetics–not a difference in what actual genes they have.
    -The twin studies account for ALL genes, no matter how small their effect and no matter whether they’re known to scientists or not. There could be a whole bunch of genes with small effects that add up to the 70-80%. (Regulating fat/weight might not be their most important function–see “pleiotropy”.)
    -Marion Nestle is then also talking about variation BETWEEN environments (America of the ’60s vs America of the 2000’s), which neither the study you link to nor the twin studies looked at. (I liked your post about your grandmother and the sugarcane for a discussion of this.) When you start comparing different environments, you’re not talking heritability any more.
    -It’s possible that some genes with a small effect by themselves could amplify the effect of other genes (or dampen the effect)

    So I’m guessing that the explanation is that there are many genes with small but real effect on weight, and/or genetic imprinting and womb environment are particularly important in affecting weight, so that although the particular genes the researchers in that study looked at had a small effect, the total biological contribution is in line with previous studies showing 70-80% heritability.

  4. November 23, 2010 8:37 pm

    Um, actually, part of that was wrong. Twin studies compare fraternal vs. identical twins, and both would share a womb environment and epigenetic effects, so a difference in the closeness of fraternal vs. identical twins argues AGAINST womb environment or epigenetics having a strong effect.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: