Xander Fulton
United States
Astoria
Oregon
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
Interesting paper published recently that points to genetics possibly playing a factor in who we are friends with

Quote:
More than any other species, humans form social ties to individuals who are neither kin nor mates, and these ties tend to be with similar people. Here, we show that this similarity extends to genotypes.


From the paper, itself:

Quote:
Pairs of friends are, on average, as genetically similar to one another as fourth cousins...


With the paper noting that this correlation does not occur between 'strangers'.

And an interesting note on where they saw some significant manifestations of this correlation occurring:

Quote:
Friend pairs tend to have similar gene variants (homophily) in the olfactory and linoleic acid systems, and different gene variants (heterophily) in the immune system. By comparison, the same analysis with stranger pairs shows no significant gene sets


1 
 Thumb up
1.00
 tip
 Hide
  • [+] Dice rolls
Moshe Callen
Israel
Jerusalem
flag msg tools
designer
ἄνδρα μοι ἔννεπε, μοῦσα, πολύτροπον, ὃς μάλα πολλὰ/ πλάγχθη, ἐπεὶ Τροίης ἱερὸν πτολίεθρον ἔπερσεν./...
badge
μῆνιν ἄειδε θεὰ Πηληϊάδεω Ἀχιλῆος/ οὐλομένην, ἣ μυρί᾽ Ἀχαιοῖς ἄλγε᾽ ἔθηκε,/...
Avatar
mbmbmbmbmb
Um, I grew up in an area of Texas where there basically weren't any other white people nor black people. I was under the impression I still had friends even though I was a freckle-faced, red-headed Jewish kid with hazel eyes and they were essentially all a deep brown with black hair and brown eyes.

EDIT:
Namely what are the qualifiers on these results?
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
United States
Boise
Idaho
flag msg tools
badge
Avatar
mbmbmb
Clearly this is science used as a thin veil to cover racism.
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Erik Henry
United States
Houston
Texas
flag msg tools
badge
Avatar
mbmbmb
whac3 wrote:
Um, I grew up in an area of Texas where there basically weren't any other white people nor black people. I was under the impression I still had friends even though I was a freckle-faced, red-headed Jewish kid with hazel eyes and they were essentially all a deep brown with black hair and brown eyes.

EDIT:
Namely what are the qualifiers on these results?

Well, they do say "on average" -- which is a pretty big qualifier, and sufficient to allow the results to still potentially be accurate even though a single exception has been pointed out.

More to the point, though, their database is mostly whites of European ancestry. So it may or may not be applicable across the board.

The findings of heterophilicity were interesting. As was the finding that friends tend to be as genetically similar as 4th cousins. Particularly when it seems like they tried to eliminate effects of population stratification, which may be what you're getting at.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Xander Fulton
United States
Astoria
Oregon
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
Erik17 wrote:
whac3 wrote:
Um, I grew up in an area of Texas where there basically weren't any other white people nor black people. I was under the impression I still had friends even though I was a freckle-faced, red-headed Jewish kid with hazel eyes and they were essentially all a deep brown with black hair and brown eyes.

EDIT:
Namely what are the qualifiers on these results?

Well, they do say "on average" -- which is a pretty big qualifier, and sufficient to allow the results to still potentially be accurate even though a single exception has been pointed out.

More to the point, though, their database is mostly whites of European ancestry. So it may or may not be applicable across the board.

The findings of heterophilicity were interesting. As was the finding that friends tend to be as genetically similar as 4th cousins. Particularly when it seems like they tried to eliminate effects of population stratification, which may be what you're getting at.


Indeed, that was a curious point.

I mean, how could anyone actually know that they had substantial variation in genetic makeup in their immune system from someone else? Must be some kind of queues we are picking up subconsciously, which is...weird.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Erik Henry
United States
Houston
Texas
flag msg tools
badge
Avatar
mbmbmb
XanderF wrote:
From the paper, itself:

Quote:
Pairs of friends are, on average, as genetically similar to one another as fourth cousins...


So in Connecticut* you're likely more genetically similar to your wife than to your friends? whistle



* I was going to use one of the more stereotypical states here, but the actual list of places where marrying your first cousin is legal is: Alaska, Alabama, California, Colorado, Connecticut, Florida, Georgia, Hawaii, Maryland, Massachusetts, New Jersey, New Mexico, New York, Rhode Island, South Carolina, Tennessee, Vermont, Virginia, and Washington D.C.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
いい竹やぶだ!

South Euclid
Ohio
msg tools
Avatar
mbmbmb
There are several features of this study that lead me to doubt its validity.

1. The authors define the kinship coefficient as the probability that alleles sampled at random from two individuals are "identical by state." Actually it's the probability that randomly sampled alleles are identical by descent. That's what the software they used measures. I'm guessing they know the difference, and hope that was just a typo.

2. The manner in which the authors select friend and stranger pairs is suspicious, given their goal. To find stranger pairs, they (p. 5) remove all pairs with social interactions, and all kin pairs. There is no better way of ensuring that friends will be more closely related, on average, than to deliberately omit relative pairs among the strangers.

3. The definition of "friends" was chosen for convenience: the Framingham data set included a question about naming a close friend (“please tell us the name of a close friend, to whom you are not related... [with whom] you are close enough that they would know where you are if we can’t find you”; Suppl. Info., p. 2). I would expect a definition of friendship on which an entire study is based to be much more careful. Could spouses count as friends?

4. A true replication study wasn't performed; instead, the authors split the Framingham data set in two, and merely repeated the analysis on the second subset. The purpose of a replication study is to confirm an initial finding in an independent sample, presumably lacking any strange confounding factors that may have generated a false positive result in the original sample. The Framingham Study is based on the population of a single, mid-sized town in Massachusetts, and so there is every reason to believe that the two subsamples are highly correlated with respect to anything that might bias the results.

5. The quote "Pairs of friends, are, on average as genetically similar to one another as fourth cousins" is not accurate; the difference in relatedness between friend and stranger pairs was about equivalent to a fourth-cousin relationship (p. 5). This also brings up the issue of background relatedness. Choose any two Caucasian Americans at random, and on average they'll be about fifth cousins. The method used to estimate kinship must have corrected out this background, but I didn't see any mention of it.

Anyway, I'm not a population geneticist, and so I can't provide as thorough a critique as would be necessary before publication. I suspect that if this manuscript were sent to a prominent journal (e.g., Nature Genetics or the American Journal of Human Genetics), it would be torn to shreds.
17 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Erik Henry
United States
Houston
Texas
flag msg tools
badge
Avatar
mbmbmb
XanderF wrote:
Erik17 wrote:
whac3 wrote:
Um, I grew up in an area of Texas where there basically weren't any other white people nor black people. I was under the impression I still had friends even though I was a freckle-faced, red-headed Jewish kid with hazel eyes and they were essentially all a deep brown with black hair and brown eyes.

EDIT:
Namely what are the qualifiers on these results?

Well, they do say "on average" -- which is a pretty big qualifier, and sufficient to allow the results to still potentially be accurate even though a single exception has been pointed out.

More to the point, though, their database is mostly whites of European ancestry. So it may or may not be applicable across the board.

The findings of heterophilicity were interesting. As was the finding that friends tend to be as genetically similar as 4th cousins. Particularly when it seems like they tried to eliminate effects of population stratification, which may be what you're getting at.


Indeed, that was a curious point.

I mean, how could anyone actually know that they had substantial variation in genetic makeup in their immune system from someone else? Must be some kind of queues we are picking up subconsciously, which is...weird.

Yeah, that does seem to be the implication . . . and you're right, that's very weird (and neat).
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Xander Fulton
United States
Astoria
Oregon
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
robigo wrote:
Anyway, I'm not a population geneticist, and so I can't provide as thorough a critique as would be necessary before publication. I suspect that if this manuscript were sent to a prominent journal (e.g., Nature Genetics or the American Journal of Human Genetics), it would be torn to shreds.


It certainly does seem like it needs some additional work - and I think they seem aware of that. I caught your question on spouses, too, which they seemed to note was worth more research...from the article (pg 15):

Quote:
It may be possible to use an approach similar to that outlined here, but with much larger samples of friendship pairs, and perhaps coupled with the addition of an equally large number of spousal pairs, to identify the genetic basis of kin detection. The extent to which friends and spouses resemble each other could itself be taken as a phenotype, and one could imagine doing a GWAS to isolate which regions of the genome contribute to our ability to pick suitable friends and spouses.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Erik Henry
United States
Houston
Texas
flag msg tools
badge
Avatar
mbmbmb
robigo wrote:
There are several features of this study that lead me to doubt its validity.

2. The manner in which the authors select friend and stranger pairs is suspicious, given their goal. To find stranger pairs, they (p. 5) remove all pairs with social interactions, and all kin pairs. There is no better way of ensuring that friends will be more closely related, on average, than to deliberately omit relative pairs among the strangers.

They do the same elimination for the friend pairs (i.e., just using ones where kinship is less than or equal to zero). Does that address your concern?

robigo wrote:
5. The quote "Pairs of friends, are, on average as genetically similar to one another as fourth cousins" is not accurate; the difference in relatedness between friend and stranger pairs was about equivalent to a fourth-cousin relationship (p. 5). This also brings up the issue of background relatedness. Choose any two Caucasian Americans at random, and on average they'll be about fifth cousins. The method used to estimate kinship must have corrected out this background, but I didn't see any mention of it.

I didn't see it mentioned either, but they do say "Pairs of friends are, on average, as genetically similar to one another as fourth cousins, which seems noteworthy since this estimate is net of mean ancestry and background relatedness." So it sounds like they've tried to correct for that?

robigo wrote:
Anyway, I'm not a population geneticist, and so I can't provide as thorough a critique as would be necessary before publication. I suspect that if this manuscript were sent to a prominent journal (e.g., Nature Genetics or the American Journal of Human Genetics), it would be torn to shreds.

And I know even less (much less) about this field. But it sounded interesting.




[I realized after three attempts that trying to use a less-than-or-equal-to sign was messing up this post. It looked fine in preview but everything after the symbol disappeared upon posting. Sorry to keep deleting my previous posts.]

[And if this one disappears too, I'm giving up.]
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jonny Lawless
United States
Bountiful
Utah
flag msg tools
Love my girls!
badge
My name is Glenn! Long have I carried Cyrus's hopes and dreams, and now I bear the Masamune as well! Henceforth, I claim them as my own! I shall slay the Fiendlord Magus and restore our honor!
Avatar
mbmbmbmbmb
whac3 wrote:
Um, I grew up in an area of Texas where there basically weren't any other white people nor black people. I was under the impression I still had friends even though I was a freckle-faced, red-headed Jewish kid with hazel eyes and they were essentially all a deep brown with black hair and brown eyes.


Well... this is awkward...
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
United States
flag msg tools
badge
Avatar
mbmbmbmbmb
The authors of this paper have been shown to consistently misuse statistical models to the point of in some of their prior results the correct application of the modeling techniques they would use would lead to results that are opposite of what they report (I believe it was the effect of friendship on obesity).

Traditionally, they have had a poor track record of properly understanding how to model network data to account for the dependencies; however, criticism of their techniques has been difficult to publish due to their high profile in terms of funding and publication outlet (Science, Nature, etc.).

In my mind, this illustrates the need to have expert methodologists as reviewers on every journal article that uses sophisticated techniques. However, having reviewed for over 50 journals, as well as serving on the Editorial Board of a number of others, you can rest assured that this is not the case.

Anyway, I would take it all with a grain of salt...
4 
 Thumb up
0.01
 tip
 Hide
  • [+] Dice rolls
CHAPEL
United States
Round Rock
Texas
flag msg tools
badge
"that's a smith and wesson, and you've had your six"
Avatar
mbmbmbmbmb
Interesting, the writer of this paper was named Times top 100 most influential people in 2009.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Moshe Callen
Israel
Jerusalem
flag msg tools
designer
ἄνδρα μοι ἔννεπε, μοῦσα, πολύτροπον, ὃς μάλα πολλὰ/ πλάγχθη, ἐπεὶ Τροίης ἱερὸν πτολίεθρον ἔπερσεν./...
badge
μῆνιν ἄειδε θεὰ Πηληϊάδεω Ἀχιλῆος/ οὐλομένην, ἣ μυρί᾽ Ἀχαιοῖς ἄλγε᾽ ἔθηκε,/...
Avatar
mbmbmbmbmb
MWChapel wrote:
Interesting, the writer of this paper was named Times top 100 most influential people in 2009.

Given SG's comment above, that oughtn't be a surprise.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Erik Henry
United States
Houston
Texas
flag msg tools
badge
Avatar
mbmbmb
SpaceGhost wrote:
The authors of this paper have been shown to consistently misuse statistical models to the point of in some of their prior results the correct application of the modeling techniques they would use would lead to results that are opposite of what they report (I believe it was the effect of friendship on obesity).

Traditionally, they have had a poor track record of properly understanding how to model network data to account for the dependencies; however, criticism of their techniques has been difficult to publish due to their high profile in terms of funding and publication outlet (Science, Nature, etc.).

In my mind, this illustrates the need to have expert methodologists as reviewers on every journal article that uses sophisticated techniques. However, having reviewed for over 50 journals, as well as serving on the Editorial Board of a number of others, you can rest assured that this is not the case.

Anyway, I would take it all with a grain of salt...


Wow, thanks.

Here's a published article criticizing these authors:
arxiv.org/pdf/1007.2876.pdf

And some related criticisms:
http://themonkeycage.org/2011/06/10/1-lyonss-statistical-cri...
http://www.slate.com/articles/health_and_science/science/201...

with further links in the Slate article . . .
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.