Hacker News Reader

Comment by blacksmith_tb | Hacker News Reader

The article doesn't touch on X-chromosome Inactivation[1] which makes all female animals mosaic. Though in the specific case in the article SCN5A isn't on the X[2].

1: https://en.wikipedia.org/wiki/X-inactivation

2: https://ghr.nlm.nih.gov/gene/SCN5A#location

View on HN

X-chromosome inactivation is a type of an epigenetic change. It is one of many that happen throughout the body - I would say there are very few cells in our bodies that have the same epigenetic markup. We've known about this for a while.

The article discusses genetic post-meiotic changes that propagate differentially through the body. We knew it happened for a while now too, but it was always presented as a curiosity or an anomaly. We are now learning that it is much more common than we anticipated. Like presented in the article, it might actually explain some clinical phenotypes as well.

Male birth order-linked gayness probability is epigenetics in the mother IIRC.

Interesting.

https://en.wikipedia.org/wiki/Fraternal_birth_order_and_male...

"Secondly, the fraternal birth order effect operates through a biological mechanism during prenatal life, not during childhood or adolescence. Direct evidence for this is the fact that the fraternal birth order effect has been found even in males not raised with their biological brothers – it has been determined that biological brothers increase the odds of homosexuality in later-born males even if they were reared in different households, whereas non-biological siblings, such as step-brothers or adopted brothers, have no effect on male sexual orientation."

That is consistency with a "kin-selection" theory of homosexuality, that within a family, evolutionary factors favor having partially non-reproductive members in a family (perhaps because in a family with two males, on average, the total reproductive fitness is greater where there is one father and one uncle, not two fathers)

The math is pretty clear: the chances that homosexuality, as a trait in and of itself, is selected for is highly unlikely. A gay person would have to result in their siblings having 2x as many children as they would have themselves, which is possible, but a really steep hill to climb. The explanation that makes much more sense to me is that homosexuality is an occasional side effect of something else which is highly adaptive to non-gay siblings who possess it.

You are assuming that for beneficial genes to pass on, it must result in an 2x number of children per generation. That assumption is false, even a 1.1x increase in children can be compounded. Also, imagine if there was a severe bottleneck event in the human population, where sub-populations with homosexuals had a much better chance of survival.

Furthermore, what math is so "clear", that you completely gloss over it?

I mistyped.

> A gay person would have to result in their siblings having 2x as many children

Should have been:

> A gay person would have to result in their siblings having 2x as many additional children

Here's the math:

An individual's children have 50% of their genes. Their nieces and nephews share 25% of their genes. Therefore, if the hypothesis is that homosexuality evolved to promote the production of nieces and nephews at the expense of producing direct offspring, this adaptation would need to result in at least two nieces/nephews for each offspring that the homosexual individual would have produced directly.

So, if we assume a stable population where each couple produces 2 offspring, a gay person would be giving up producing 2 of their own children. In order for this strategy to break even, it would have to result in their siblings producing 4 offspring, in addition to the offspring that they would have had anyway.

Possible, but not likely.

On top of the math just being heavily stacked against this being an explanation, there are other reasons why it is unlikely:

1. If this were a successful strategy, you would expect to see a much higher prevalence of gay people than we currently do (<5% of the population).

2. There are already adults who take on this hypothetical role of assistant to the breeders: grandmothers. And they avoid the fertility tradeoff because they are evolved to stop being fertile at a certain point in their lives.

Humans have many traits that seem to support survival at the tribal level, and having a spectrum of homosexual and heterosexual individuals could well have supported child rearing or reduced divisive competition. Just look at bonobos.

See my reply to rubatuga. If the genetic math doesn't work out for helping raise your sibling's children vs. having children of your own, it absolutely doesn't work for raising the tribe's children (who are less related to you than your sibling's children) vs. having your own.

I think you missed my point. If survival of human groups generally occurred at the tribal level, i.e. either a tribe successfully cooperated and survived or suffered excessive discord and died out unless they were adopted by a different tribe, the benefits would not rely just on direct actions within an atomic family unit. The group had to have enough genetic diversity, or invite enough in, to prevent disease from inbreeding but generally most of the tribe would be somewhat related. The "math" surrounding just caring about whoever has 1/2 or 1/4 of one's genetic material may not be as important as the survival of the larger cooperative group. All of those little selfish increments might not matter if isolated, catastrophic group collapse rendered them nil. I think you are trying to over-simplify the scenario. Other traits point to increased cooperation being crucial, such as the idea that concealed ovulation attenuated direct male competition during fertile periods. Even in The Selfish Gene, Dawkins describes game theory situations where certain ratios of populations are stable, like the simplified hawks versus doves. In the same way, there may have been a stable ratio of homosexual tendencies or sexual flexibility that acted as a cooperative buffer, as they would assist in child rearing in the same way that you mention grandmothers. Trying to boil it down to just grandmothers because they might have the largest genetic "interest" is overly reductionist. It's not just an addition problem. And bonobos are a great example of all with all sexual bonding being very close to us genetically. Who's to say that our cultural notions of what homosexuality entails translates across to thousands of years of pre-historical culture?

Don’t organs, limbs, spots and stripes form in a fetus due to epigenetics?

You're not wrong, it's a mixture of epigenetic changes, and local environment changes. As a cell differentiates, it definitely changes its epigenome. These changes are also mediated by cell-cell interactions, and other morphogenic proteins, for example, the SHH gene: https://en.wikipedia.org/wiki/Sonic_hedgehog

No, they form because of genetics: differential gene regulation.

> which makes all female animals mosaic

Nitpicking: We have the XY system but birds have the ZW system where the male birds have the chromosomes ZZ and the female birds have the chromosomes ZW (More info: https://en.wikipedia.org/wiki/Sex-determination_system )

I still can't find out if in this case the male birds have a random inactive Z and are mosaic. I guess that there is something similar, but I'm not sure.

Thanks, I was just lifting that from the wiki article (arthropods are animals too, and they don't have XY chromosomes either, of course). Those crazy feathered therapods! Plus, nucleated red blood cells[1].

1: http://www.askabiologist.org.uk/answers/viewtopic.php?id=929...

IIRC not marsupials, which get the same chromosome chosen over the entire body.

Reply on news.ycombinator.com