quote:
Originally posted by unionist:
[b]No, of course I'm not. But could you ask her to phone me when she has a minute?[/b]
Sorry, dude, she's out with her friends EFA, Chilly(ie) and God knows who else.
very basic genetics question
Mon, 2007-05-28 21:23
#102
Mon, 2007-05-28 21:24
#103
Sven
Whoops!! How the heck did that double post occur???
[ 28 May 2007: Message edited by: Sven ]
Mon, 2007-05-28 21:25
#104
I think you already posted that, Sven.
Mon, 2007-05-28 21:26
#105
You can say that again.
Mon, 2007-05-28 21:29
#106
quote:
Originally posted by Sven:
[b]A study was conducted at the University of Minnesota of identical twins separated at birth to examine the link between genes and homosexuality. Although if one twin was gay, it wasn't a 100% guarantee that the other twin would be gay, but the study concluded that there was a significant correlation, thus showing a genetic component for homosexuality.
I agree that understanding environmental factors that influence crime would be more productive, but the study of a genetic component would be interesting from a perspective of moral culpability (i.e., is a person less morally culpable for behavior if the behavior is, in part, based on genes?).[/b]
Sven,
Twins share genes.
They also share the same embryonic environment.
That must be very important in terms of sexual development.
Mon, 2007-05-28 21:30
#107
Just wanted to clarify a statement which could easily be misinterpreted:
[b]Nah, we don't want to go there. It won't be about making someone less culpable. It'll be about genetic discrimination prior to a crime that may never have happened anyway.[/b]
I'm saying a hypothetical genetic correlation to crime won't be used for a good purpose. It'll only be used for a bad purpose -- like discrimination on the basis of "pre-crime," like certain racial groups being barred from shopping malls.
Don't want to be misunderstood on my last (ha!) thread.
Mon, 2007-05-28 21:33
#108
quote:
Originally posted by 500_Apples:
[b]They also share the same embryonic environment.That must be very important in terms of sexual development.[/b]
I guess a true genetic study would actually separate identical twins right after conception.
Mon, 2007-05-28 21:37
#109
quote:
Originally posted by Westerly:
[b]I guess a true genetic study would actually separate identical twins right after conception.[/b]
Clone studies are a theoretical subject.
Or you could compare fraternal twins and identical twins separated at birth.
One problem among among many, because identical twins are so similar, they'll get a similar environment, but then you don't know where outcome similarities come from.
The problem is very nontrivial, and I think Sven's assumption is naive.
Mon, 2007-05-28 21:41
#110
quote:
Originally posted by 500_Apples:
[b]Or you could compare fraternal twins and identical twins separated at birth.[/b]
Yeah, this was discussed earlier in the thread. But, as you pointed out, fraternal twins are only 50% related on average. It's possible, although freakishly unlikely, that fraternal twins could be 100% related, right?
Mon, 2007-05-28 21:44
#111
quote:
Originally posted by Westerly:
[b]Yeah, this was discussed earlier in the thread. But, as you pointed out, fraternal twins are only 50% related on average. It's possible, although freakishly unlikely, that fraternal twins could be 100% related, right?[/b]
If they're the same gender, about one in four trillion. Which means it has likely never happened.
Mon, 2007-05-28 21:45
#112
Sven
quote:
Originally posted by unionist:
[b]How can you ask such a question when there's no shred of evidence of such a link?[/b]
For Christ's sake, unionist, would you please read my posts??? I was asking earlier if such a study should even be made (to see [b][i]whether[/b][/i] there is a link). I then stated that such a study would be interesting, because if there was a link it would raise an interesting question about moral culpability. But...
You then jumped all over and shouted, "There's no shred of evidence of such a link"!!!
That's not even what I'm saying.
Jezuz. Learn to read before jumping on someone. 'Kay?
[ 28 May 2007: Message edited by: Sven ]
Mon, 2007-05-28 21:46
#113
quote:
Originally posted by 500_Apples:
[b]If they're the same gender, about one in four trillion. Which means it has likely never happened.[/b]
Gotcha. But they could be many other percentages other than 50, no? I'm just saying it's hard to count things when you don't really know what you're counting.
Mon, 2007-05-28 21:50
#114
quote:
Originally posted by Sven:
[b]For Christ's sake, unionist, would you please read my posts??? I was asking earlier if such a study should even be made (to see [b][i]whether[/b][/i] there is a link. I then stated that such a study would be interesting, because if there was a link it would raise an interesting question about moral culpability. But...You then just all over and shout, "There's no shred of evidence of such a link"!!!
That's not even what I'm saying.
Jezuz. Learn to read before jumping on someone. 'Kay?[/b]
Sven, there have been such studies done -- the criminal body type theory, etc. They failed to produce any evidence of a genetic link to criminality. Now I don't know that there will never be similar studies in the future as gene sequencing, etc. progresses. That's why I think the subject should be discussed ahead of time. And, for those about to jump all over me, the "subject" to which I refer is whether such studies are ethical.
Mon, 2007-05-28 21:52
#115
Sven
quote:
Originally posted by 500_Apples:
[b]The problem is very nontrivial, and I think Sven's assumption is naive.[/b]
What assumption?
Mon, 2007-05-28 21:54
#116
quote:
Originally posted by Sven:
[b]What assumption?[/b]
I think he means that identical twins raised apart may not be a true comparison to regular identical twins.
Mon, 2007-05-28 21:55
#117
quote:
Originally posted by Westerly:
[b]Gotcha. But they could be many other percentages other than 50, no? I'm just saying it's hard to count things when you don't really know what you're counting.[/b]
Math Games.
There's 22 chromosomes pairs, plus the sexual pair.
1) Assume no mutations (on average there's a few per generation)
2) ignore the sex chromosomes for now.
3) For each chromosome pair of 1 through 22, you have four choices, as you get 1 each from mother and father.
So what are the odds of having 22 of 44 chromosomes in common (exactly 50% among the non-sexual ones)?
44C22 * (1/2)^44 *100% = 11.9% chance of sharing exactly 22 of them.
I didn't weigh for the fact some chromosomes are more important than others, as I don't know which are more important.
For comparison, what are the odds of sharing 30 of 44?
44C30 * (1/2)^44 *100% = 0.65%
Or what about 26/44?
it's 5.8%.
So while the probably of exactly 50% split is surprisingly small, it drops off very fast.
Mon, 2007-05-28 21:57
#118
quote:
Originally posted by Sven:
[b]What assumption?[/b]
On the ideal of twins as a control study of genetics versus environment.
Twins share genetic emvironment and embryonic environment, and so if they share more in common at age 25, how do you decide it's one and not the other?
Your post, as I read it, implied "genes".
You COULD be right.
I just don't think it's obvious.
Mon, 2007-05-28 22:03
#119
Sven
quote:
Originally posted by 500_Apples:
[b]On the ideal of twins as a control study of genetics versus environment.
Twins share genetic emvironment and embryonic environment, and so if they share more in common at age 25, how do you decide it's one and not the other?
Your post, as I read it, implied "genes".
You COULD be right.
I just don't think it's obvious.[/b]
So, what better model would you use to replace the study of identical twins separated at birth?
Mon, 2007-05-28 22:07
#120
quote:
Originally posted by 500_Apples:
[b]So while the probably of exactly 50% split is surprisingly small, it drops off very fast.[/b]
I trust your math but that result is not intuitively obvious, to say the least.
Mon, 2007-05-28 22:08
#121
quote:
Originally posted by Sven:
[b]So, what better model would you use to replace the study of identical twins separated at birth?[/b]
I'm not saying I have a better model, I'm saying the best model I'm aware of is nonperfect. I have some common sense and some math but I don't specialize in the field. Also, are there really that many twins separated at birth that we have kept track of?
A friend of mine who did psychology told me clone studies are a much discussed potential for the future. Clones will be raised in a different embryonic environment and grow up in a different culture. Additionally, with advances in genomics we'll be able to pinpoint individual genes and geneplexes a lot more, and as we get more precise data we might also get more targetted observations.
Don't get me wrong, I think the studies of twins separated at birth are very powerful. However, how we interpret the power in the data does not strike me as self-evident.
Mon, 2007-05-28 22:10
#122
Sven
quote:
Originally posted by 500_Apples:
[b]I'm not saying I have a better model, I'm saying the best model I'm aware of is nonperfect.[/b]
Who the fuck said it's a perfect model?
Mon, 2007-05-28 22:13
#123
quote:
Originally posted by 500_Apples:
[b]A friend of mine who did psychology told me clone studies are a much discussed potential for the future. Clones will be raised in a different embryonic environment and grow up in a different culture. [/b]
This might be a bit icky. What if the clone grows up and doesn't want to participate in a study? And wouldn't those clones be bred specifically for experimentation? How is that going to fly? Seems like an ethical dilemma for sure.
Mon, 2007-05-28 22:16
#124
quote:
Originally posted by Westerly:
[b]I trust your math but that result is not intuitively obvious, to say the least.[/b]
The way to check your answer in probability is to repeat the experiment with smaller numbers and see if you get the same trends. Imagine humans had 3 chromosomes from dad and 3 from mom. Write down all the ways to get 3/6 shared, and all the ways to share 2/6 and 1/6, you'll see a fast dropoff.
Anyhow, my math may be confusing your intuitoon, because you may be thinking some siblings look a lot more alike than other pairs of siblings. And that's because I didn't account for the fact not all chromosomes are equal.
Mon, 2007-05-28 22:18
#125
quote:
Originally posted by Sven:
[b]Who the fuck said it's a perfect model?[/b]
1) calm down.
2) you wrote:
quote:
A study was conducted at the University of Minnesota of identical twins separated at birth to examine the link between genes and homosexuality. Although if one twin was gay, it wasn't a 100% guarantee that the other twin would be gay, but the study concluded that there was a significant correlation, thus showing a genetic component for homosexuality.
It's the last sentence that bothered me.
Mon, 2007-05-28 22:19
#126
Sven
quote:
Originally posted by 500_Apples:
[b]It's the last sentence that bothered me.[/b]
That there is a "signficant correlation"? What bothered you about that?
Mon, 2007-05-28 22:21
#127
quote:
Originally posted by Sven:
[b]That there is a "signficant correlation"? What bothered you about that?[/b]
It said: "thus showing a genetic component for homosexuality."
How do they know the significant correlation is not due to embryonic environment?
Have you been reading what I've been writing Sven? I think most of my basic points were clear. Westerly got them. I have no desire to fight you at 2:21am.
Mon, 2007-05-28 22:23
#128
quote:
Originally posted by 500_Apples:
[b]The way to check your answer in probability is to repeat the experiment with smaller numbers and see if you get the same trends. Imagine humans had 3 chromosomes from dad and 3 from mom. Write down all the ways to get 3/6 shared, and all the ways to share 2/6 and 1/6, you'll see a fast dropoff.Anyhow, my math may be confusing your intuitoon, because you may be thinking some siblings look a lot more alike than other pairs of siblings. And that's because I didn't account for the fact not all chromosomes are equal.[/b]
I know I should figure this out for myself but shouldn't all the probabilities (1/44 = X%, 2/44 = Y%, etc.) add up to 100%? If the 50% mark is only around 11%, I don't see how they could add up to 100% when the, for e.g., 30/44 percentage likelihood is so low.
Not all chromosomes equal. Not sure what this means, although I do know that Mendel's AB/CD scenario only works for some simple hereditary situations (apparently, he fudged a little bit to make the pea experiments work so nicely).
Mon, 2007-05-28 22:27
#129
Sven
quote:
Originally posted by 500_Apples:
[b]It said: "thus showing a genetic component for homosexuality."
How do they know the significant correlation is not due to embryonic environment?
Have you been reading what I've been writing Sven? I think most of my basic points were clear. Westerly got them. I have no desire to fight you at 2:21am.[/b]
I get it. You don't believe there's genetic component for homosexuality.
Mon, 2007-05-28 22:29
#130
quote:
Originally posted by Westerly:
[b]I know I should figure this out for myself but shouldn't all the probabilities (1/44 = X%, 2/44 = Y%, etc.) add up to 100%? If the 50% mark is only around 11%, I don't see how they could add up to 100% when the, for e.g., 30/44 percentage likelihood is so low.
Not all chromosomes equal. Not sure what this means, although I do know that Mendel's AB/CD scenario only works for some simple hereditary situations (apparently, he fudged a little bit to make the pea experiments work so nicely).[/b]
1)
Sharing 22 of 44 non sexual chromosomes has odds of 11.9%. Sharing either 21 or 23 has odds of 11.4%. 20 or 24, 10.1%. 19 or 25, 8%. 18 or 26, 5.8%. 17 or 27, 3.9%.
So for between 17/44 and 27/44, that's around 90%.
2) There should be as many sibling pairs who share chromosome 1 pair but not 22, as there are pairs who share chromosome 22 pair but not chromosome 1. But those who have the same lineup of chrosomome 1 should have a lot more in common.
(I think Wikipedia says that for the non-sexual chromosomes, smaller numbers means more genes).
Mon, 2007-05-28 22:31
#131
quote:
Originally posted by Sven:
[b]I get it. You don't believe there's genetic component for homosexuality.[/b]
No way, Sven. That's not what 500_Apples is saying at all. He's saying that the correlation which you discuss (identical twins separated at birth vs. regular identical twins) isn't a slam dunk for genetics over environment. The embryonic environment (which is shared) is also an environmental issue, i.e. non-genetic.
Mon, 2007-05-28 22:31
#132
quote:
Originally posted by Sven:
[b]I get it. You don't believe there's genetic component for homosexuality.[/b]
No, I said I'm not convinced that there is.
Mon, 2007-05-28 22:35
#133
quote:
Originally posted by 500_Apples:
[b]So for between 17/44 and 27/44, that's around 90%.
[/b]
Wow. Yeah, that is eaten up pretty quickly. I'm not a good estimator.
quote:
[b]There should be as many sibling pairs who share chromosome 1 pair but not 22, as there are pairs who share chromosome 22 pair but not chromosome 1. But those who have the same lineup of chrosomome 1 should have a lot more in common.[/b]
Yeah, I see what you mean now. Chromosome 1 has more effect because it has more genes. Well, shit, then we should be dealing with genes, not chromosomes. Will that ever be possible?
Mon, 2007-05-28 22:40
#134
quote:
Originally posted by Westerly:
[b]Yeah, I see what you mean now. Chromosome 1 has more effect because it has more genes. Well, shit, then we should be dealing with genes, not chromosomes. Will that ever be possible?[/b]
I was just playing a probabilistic game with the chromosome as a qualitiative way of seeing how related siblings are.
I think the scientific process is first to observe generalized correlation, then to try and reduce the correlation to a single gene or a cluster of genes, if possible. And then that's hard as well, bceuase some genes are off and some on, and gene-environment interactions, et cetera. Maybe in a year or two phrillie, who will then be an expert on genetics, can bump this thread and share some of her knoweldge.
Good night. I predict Michelle or Oldgoat will close this thread in the morning because it's ove 100 posts.
Mon, 2007-05-28 22:44
#135
quote:
Originally posted by 500_Apples:
[b]Maybe in a year or two phrillie, who will then be an expert on genetics, can bump this thread and share some of her knoweldge.[/b]
[img]smile.gif" border="0[/img] Awesome, thank you. And thanks for all the info tonight.
Mon, 2007-05-28 23:02
#136
quote:
Do you honestly think a newborn infant of Jewish parents, raised from birth by a (say) Lutheran couple with no knowledge of the baby's origins, would grow up being able to deliver a decent punch-line?
You're saying Lutherans are an ethnic group? Or are you saying Jews are defined solely by religion (so that you can point out there's probably not a gene for chosenness)? Or are you really saying something else?
quote:
How can you ask such a question [re the genetic determination, to at least some degree, of behaviour] when there's no shred of evidence of such a link?
That McCoy reference upthread, then, you would say is not such a shred?
You seem somehow threatened by the idea of genetic determinism. But it's only a world-view, you can get another one tomorrow.
Pages
- « first
- ‹ previous
- 1
- 2
- 3