PDA

View Full Version : On Genetics



Stormcrow
2007-11-06, 08:53 PM
Oki Dokie. To move away slightly from the usual;

I'm working on a campaign world and my question is thus, can you take four families of 50-100 members each and have them breed nigh exclusively and have each family survive viably. Or is one hundred not enough people to make each family genetically different enough.

In summary.
Each clan is genetically pure. Just that clan's bloodline, is 100 people enough to start that process?

<If you don't know genetics, please don't comment>

Iku Rex
2007-11-06, 09:00 PM
Each clan is genetically pure. Huh? What does "genetically pure" mean?

Stormcrow
2007-11-06, 09:18 PM
To explain more clearly.
Say an extended family of one hundred walks into a forrest together, are they enough building blocks for a society to flourish there, genetically. Without people from outside of that original pool coming in.

Mewtarthio
2007-11-06, 09:19 PM
I believe he means that nobody has any significant genetic disorders that would preclude breeding.

horseboy
2007-11-06, 09:25 PM
I think it's 50/500
50 individuals for short term survival
500 individuals for long term viability.

Prometheus
2007-11-06, 09:27 PM
Other than the societal taboo of inbreeding, I don't think medieval people knew of the negative genetic effects of inbreeding. Therefore as a DM you can just say that there are no negative genetic effects or that genetics doesn't work the same way.

Probably the better question you want to ask is if the PCs will see it as distasteful inbreeding, and the answer is probably not exclusively, but it will make them a little uneasy. If that is the effect you are going for then you've got it perfect. If you want no element of disgust, go with 1000, if you want a high level of disgust, go with 30.

Rex Blunder
2007-11-06, 09:29 PM
Wonder which genetics thread is the fittest :smallwink:

What's the scenario? Does each family breed with members of the other families, or amongst itself? If the latter, what's the definition of "family"?

Pitcairn Island started with 25 unrelated people. After 70 years, their population was 200, and I assume that at that point, everyone was related to each other. Their colony has kept going until today without much new blood. It's a good way to collect genetic defects though.

Stormcrow
2007-11-06, 09:29 PM
I've got a cultural anthropologist in my gaming group so I try to get stuff right as often as I can.

Rex Blunder
2007-11-06, 09:33 PM
Any chance of letting the cultural anthropologist design some of the campaign world? It'd give him/her some ownership and give you a well-researched world. Of course, if you need to spring surprises on the party, that might not work.

Lord Tataraus
2007-11-06, 09:35 PM
If it is a Lawful group (alignment-wise), I suggest doing what the aborigines of Australia did which is ingenious and I'll ways remember it. First, its been awhile since I went Down Under and learned this, so this is the best I can remember:

The whole group splits into 8 different subgroups A-H.
Males stay in their subgroup, females are married out.
Females from group A marry into group B, females from group B marry into group C, etc.
So basically you get something like this:
A -> B -> C -> D -> E -> F -> G -> H -> A...

Due to this organization, the small isolated groups of aborigines survived without any genetic disorders with relatively small numbers.

Xefas
2007-11-06, 09:36 PM
And this anthropologist isn't more concerned about the fact that every living thing in D&D can be half something else, even if they don't even use the same breeding method? or are even of comparable size?

Where do the half-dragon badgers come from? Where?!

BizzaroStormy
2007-11-06, 09:36 PM
First of all, are you wanting some specific trait to be passed on to every member?

Stormcrow
2007-11-06, 09:39 PM
You did catch I'm... _from_ Australia yeah?

But I'm not touching that one with a thousand yard pole.

Basically the idea is that having rifted from their society the four effected families move to a provence called Voetaari and there is some issues with locals etc but basically each family moves into a different area of the country and venerates a different nature spirit that dwells there.

The familes of 100ish people mostly breed amongst themselves minimum distance of cousins however, and ocasionally other tribes send people for alliances and stuff. I want to know if its viable or if I need more principle people before I write their history.

StickMan
2007-11-06, 09:40 PM
I would like to note here that 25% of all marriages world wide are between cousins. I think your going to need more than 100 but it also depends on what your family set up is. If all of these people are related with in one generation your in trouble i.e. they are all first cousins, brother and sister your doomed. If there is more diversity than that such as there are 3 cousins and such in large numbers you might be OK with that small of a number.

Odds are the family would be very careful about who is married to who, even early people figured out that mating with close family is bad that is why its taboo in many cultures. When you want to marry or mate with someone your going to need to be at least second cousins for it to be OK with the clan.

This is only if your talking humans some species do better with inbreeding some do worse. Especially in a DND world.


Well this is large than 100 people but look how the European royalty look. Waits to be killed by European Ninjas.

HidaTsuzua
2007-11-06, 09:47 PM
200-400 people is likely enough to keep a population going. Minimum Population Size is based off of probability of extinction at that size (usually 5-10%), not a guarantee. Polynesia studies and theories are a good source of ideas about colonization. Here's a paper that gives around ~70 women for the founding of New Zealand (http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=21200). There are others that give as low as 20-30 people total. So it's okay especially if you up the numbers a bit. Throw in some interbreeding and things are even better.

However you will likely have a huge amount of genetic drift. With such a small population, the chances of any given allele (variations of a certain gene) becoming fixed (everyone has it) or gone (no one has it) will be great [this is the usual fate for most alleles, but the speed depends on chance and effective population size]. So people will look similar (not identical) and often have stuff rare in other populations (6 fingers, a genetic disease, the Habsburg Jaw, (http://www.antiquesatoz.com/habsburg/habsburg-jaw.htm) and more).

Lately please ditch "genetically pure." While I think I know what you're talking about, it sounds really bad. Also really isn't a factor anyways (the bottleneck of 100 people will do it).

Lord Tataraus
2007-11-06, 10:01 PM
You did catch I'm... _from_ Australia yeah?

O.o


Oh...*hides*

OneWinged4ngel
2007-11-06, 10:07 PM
As far as I know, inbreeding doesn't cause much problems in and of itself (contrary to popular belief) so much as simply causing recessive traits to show up more often (which can be beneficial OR detrimental.)

AtomicKitKat
2007-11-06, 10:15 PM
Inbreeding is both beneficial and detrimental. And it's the frequency of ocuurence of recessive traits which provides both. Assuming that a recessive trait is "obvious" and the families are very strict about enforcing "purity", they will exile the rejects into cloistered "villages"(possibly separated by gender, or neutered, or even outright killed, depending on alignment). Over time, only the fittest will survive, in much the same way that dogs have been forced into very specific breeds over millenia.

Ulzgoroth
2007-11-06, 10:16 PM
Loss of heterozygosity has reliable bad consequences. There are just so many places for it to go wrong, since it lets any defect in the ancestor get a chance at homozygosity. Yes, you might come across a useful recessive trait, but the bearer would probably be nice and messed up in all kinds of ways.

Also tends to reduce fertility, because there's a big reservoir of prenatal (often conception-level) recessive lethal alleles that get a much better chance of coming up than usual.

Chronos
2007-11-06, 10:17 PM
...so much as causing recessive traits to show up more often (which can be beneficial OR detrimental.)Usually, if the recessives are beneficial, they'll tend to become more common anyway, and if the advantage is significant, will quickly (as evolutionary timescales go) spread through the population.

Really, though, you can almost go to an arbitrarily small seed population, if you're willing to accept very high rates of defects. If you just have lots and lots of kids, and maybe cull out (by death, exile, or sterilization) the ones which show undesired traits, you'll eventually get a population without those traits. Of course, this is morally and ethically questionable for humans.

StickMan
2007-11-06, 10:26 PM
People above bring up good points. If the starting group is very healthy you should be good, but if they do have a negative trait it will become pronounced in the population. I think your safe with the group size honestly.

Rex Blunder
2007-11-06, 10:27 PM
If you just have lots and lots of kids, and maybe cull out (by death, exile, or sterilization) the ones which show undesired traits, you'll eventually get a population without those traits.

I don't know if that's true. Aren't there hereditary diseases like cystic fibrosis which are genetic but usually fatal in childhood? I think recessive genes can survive in the population even if, when they cause their trait, they prevent breeding.

Sundog
2007-11-06, 10:32 PM
I don't know if that's true. Aren't there hereditary diseases like cystic fibrosis which are genetic but usually fatal in childhood? I think recessive genes can survive in the population even if, when they cause their trait, they prevent breeding.

Actually, such recessives can only survive in an "open" populaton. In a closed population, lethal recessives are either bred out or become dominant - in which case the population crashes to either a much smaller population without the recessive, or to extinction.

Nerd-o-rama
2007-11-06, 10:39 PM
Well, the only thing I've ever heard is that you need about 200 genetically distinct men to repopulate the planet. That was with the better part of 3 billion women to breed with, however. Also it was in a comic, but Brian K. Vaughn seems to do his research.

500 sounds like a good estimate.

Dervag
2007-11-06, 10:46 PM
Pitcairn Island started with 25 unrelated people. After 70 years, their population was 200, and I assume that at that point, everyone was related to each other. Their colony has kept going until today without much new blood. It's a good way to collect genetic defects though.The Pitcairn islanders do have some issues with genetic defects, though not crippling ones.


I would like to note here that 25% of all marriages world wide are between cousins.For what value of 'cousin'?


Well this is large than 100 people but look how the European royalty look. Waits to be killed by European Ninjas.Actually, most of the European Ninjas ended up backing various republican or socialist movements, not least because the alternative was all those inbred monarchs.

On the other hand, European royalty tended to marry cousins more often than populations would indicate, because they kept trying to keep various noble titles in the family. Perversely, they actually had an incentive to engage in mild inbreeding (i.e. not brother/sister, but second cousin/second cousin or occasionally first/first).


As far as I know, inbreeding doesn't cause much problems in and of itself (contrary to popular belief) so much as simply causing recessive traits to show up more often (which can be beneficial OR detrimental.)For a group with reasonable distributions of genes, it tends to be detrimental. Many recessive gene combinations are fatal; few recessive gene combinations are so beneficial that it offsets the increased risk of death.


People above bring up good points. If the starting group is very healthy you should be good, but if they do have a negative trait it will become pronounced in the population. I think your safe with the group size honestly.The problem is that they can be perfectly healthy and still carry nasty recessive genes.


Actually, such recessives can only survive in an "open" populaton. In a closed population, lethal recessives are either bred out or become dominant - in which case the population crashes to either a much smaller population without the recessive, or to extinction.The risk of this happening is, of course, the entire problem with inbreeding as a practical matter.

Clementx
2007-11-06, 11:07 PM
For the majority of human cultures for most of human history (lots of generalization there, but it bears out), the preferred mate was the 1st cousin, usually for reasons of inheritance. Combine that with the lack of mobility and low-population density for most of history, you can see that humans do just fine in small clusters. The whole incest taboo is purely subjective. Western modern cultures think that 1st cousins are gross. Unilinear cultures would think nothing of marrying a cross-cousin, but be just as repulsed by a parallel-cousin.

One hundred gives you plenty of familiar clusters, and as long as they are grouped into moieties, exogamous clans, or other divisions, describing them as non-Dunwichean won't raise any flags for verisimilitude.

Nonah_Me
2007-11-07, 12:45 AM
I want to link this thread to the thread where the guy says most people on the boards are commoners with less than a 13 INT.

dragonseth
2007-11-07, 01:23 AM
Hooray for the founder effect! Inbreeding doesn't cause mutations by itself, but it will increase the rate of homozygosity in the population. It is important to note, though, that allelic frequencies do not change, only genotypic frequencies do. And so, individuals who are homozygous recessive for some genetic disorder become very common. A good example is the Amish. EVC syndrome and glutaric acidurea both have very high rates of occurence in some Amish communities.

On a related note: Support Hardy-Weinberg Equilibrium! Practice random mating!

Nonah_Me
2007-11-07, 01:37 AM
On a related note: Support Hardy-Weinberg Equilibrium! Practice random mating!

That, sir, is going in my sig. Thank you.

dragonseth
2007-11-07, 01:49 AM
That, sir, is going in my sig. Thank you.

Our branch of Tri-Beta sells shirts with that on them. It's really funny. They also have shirts with a cell undergoing cytokinesis that say "Stop staring at my cleavage!"

NerdVicious
2007-11-07, 04:30 PM
Alright, some people have touched on this. The short answer is there is no real way of knowing, but it most likely will NOT work.

Humans have a high "genetic load". This means that we have a lot of deleterious ("bad") recessive traits. Most of the time, we'd never know we had them. Why? We typically have another gene complimentary to the bad one that is dominate.

So what does this mean? First off, families tend to have a lot of similar genes ( I should get a lot of "Duhs" for this one). So, lets say that Cousin It and Cousin What are 2nd cousins. They both got a recessive gene from their Great-Grandpa Where. Lets call this recessive gene A1. They both also have a gene A2. Neither of them actually know this since they've never gotten genetic mapping done, and the A2 gene is dominate. That is to say that they would SEEM to be like an A2A2 person - completely healthy.

So Cousin It and Cousin What are BOTH A1A2 individuals. If they started to produce children, 25% of them would be A1A1 individuals and would die. 50% would be just like them - A1A2. 25% would be A2A2 and not carry the recessive gene at all.

What would happen after a LONG time is eventually all of those "bad" recessive genes would get bred out of a population because those with the highest fitness (not carrying the recessive gene at all) will be more likely to produce more and healthier progeny. Those who had both genes, A1A2 will probably survive to breed, but will produce less children than the A2A2 individuals.

So its not that the family doesn't have enough genetic variation. Technically, its possible that they do and with a long enough time span they could get rid of all of their negative traits.

Its the small population size that is the core issue here. With only ONE bad recessive gene in the family, 25% of the kids in the next generation will die. If there are more than just one bad recessive gene like this, that percentage continues to increase. This is just getting to the stage of being a baby. That baby has to then survive to the age of being a breeding adult (not getting killed by koblolds, falling into a pit, so on...), and then their children will also have some likelihood of dying due to residual "bad" recessive genes.

With 100 people, you're running the risk that your population would simply die in a few generations. Note I said risk - its not guaranteed. It is possible, though unlikely, that a population of 100 family members could survive despite inbreeding.

However, if it were Vegas odds, I'd bet heavily against it. Inbreeding negatively effects most populations of living organisms with the exception of those who have evolved TO inbreed (plants actually have a high rate of inbreeding, but they don't have a "genetic load" anymore).

**Edit**

Addressing the possibility of beneficial recessive alleles in the population of the family - don't bet on it. Beneficial changes like that will tend to be a VERY small number of recessive genes. Think about it. Lets take a working computer. Now lets make a random change to it - anything. This could be switching out hardware, installing software, running the computer under water, hitting the computer with a baseball bat, making a motherboard out of oreos.... I do mean "random" here. A vast majority of random changes will be bad. Some will do virtually nothing. A tiny portion of them might be good.

In other words - "Its not broken, don't fix it".

In the instance of a recessive beneficial allele cropping up, it would get noticed quickly through natural selection as soon as someone had both copies of the gene. If they only have one... oh well. Natural selection is blind and wouldn't know it was there anyway, and they would be treated just like every other individual in the population.

Long story short - the argument that beneficial mutations eventually out weighing the deleterious ones is a faulty argument.

And, yes, lets all support Hardy-Weinberg! :smallredface:

Irenaeus
2007-11-07, 05:11 PM
Also it was in a comic, but Brian K. Vaughn seems to do his research.

He also seems to manage to produce very high quality material on a regular basis. More people should read his stuff.

Stormcrow
2007-11-07, 05:53 PM
Ok. Now I'll change the details and see how we go.
Say that we concentrate on one family for now.
They used to be a part of a kingdom and we assume that their are clan lines here but that only 50 of the hundred people are genetically related and the other fifty are spouses they took while still a part of the other kingdom.

Is this more viable?

Greenfaun
2007-11-07, 06:18 PM
I think it's 50/500
50 individuals for short term survival
500 individuals for long term viability.

I have also heard this, but I don't remember from where.

Anyway, there's been some great discussion so far, I'll just mention one thing that sort of has bearing: I read an article that said that the statistics of genetic drift indicate that the ancestral line of modern humans dwindled down to something between 1,000 and 10,000 individuals, which is astonishingly small for a group that became a distinct species, according to that same article. I can't seem to find it at the moment, so I might be getting a detail wrong, but still, pretty cool.

Also, I just want to cheer a little bit for this forum. More evidence that the playground is by far the smartest gaming forum on the intarwub. Awesome.

NerdVicious
2007-11-07, 06:32 PM
I read an article that said that the statistics of genetic drift indicate that the ancestral line of modern humans dwindled down to something between 1,000 and 10,000 individuals, which is astonishingly small for a group that became a distinct species, according to that same article. I can't seem to find it at the moment, so I might be getting a detail wrong, but still, pretty cool.

I think that it was 10,000 individuals was the numbers we were at until a few thousand years ago. Its kinda astonishing to think about. :smallbiggrin: I think we have accumulated quite a lot of deleterious recessive alleles since then though.

Anyway...

There's higher genetic variation, so there is a slightly greater chance that they'll survive. However, here's the following problems:

1) They all lived in the same place, right? Thus, it follows that though they may not be closely related, there is the possibility of SOME type of recessive allele being present that is unique to that entire populace.

2) We are going off of the assumption that we start with 50 related individuals and 50 spouses (and for the hell of it, none of them are related to each other OR to our 50 person family). We are also making the assumption that no new individuals enter into the populace at any given point in time. Thus, after only a few generations, inbreeding will inevitably start up. Its effect will not be as pronounced, but it will still follow a similar pattern as stated above.

3) Its still a VERY small population size, even for humans. Someone else mentioned the idea of "genetic drift" earlier. The concept is that lets say we got 2 alleles. Eventually, in any population, one allele WILL be lost and the other will be fixed.

Here's an illustration of the idea. Lets say we have moths that are white and moths that are black. They are in a black setting, so the white moths are more likely to die and less likely to breed. Over time, the genes conferring the white trait run a high risk of being lost.

I know that sounds a lot like natural selection, but its not. :\ I'm just having a hard time coming up with a good example.

Anyway, so eventually, your entire populace, over time, will all carry pretty much the same alleles. Put another way, eventually they will all be related. Each generation that passes increases this likelihood.

Final point: It could work, but I still think its not that likely. If they did some how survive the bottleneck, they'd probably all be, well... the victims of several mental and physical afflictions even if they DID survive, and the chances that they would still be "normal" (similar to a randomly chosen person from a non-inbred populace) would be low.

If you still want to go with the idea, I don't think that the effects of inbreeding would be as high if it was within the first 4 generations (not including the initial generation). Past that, it starts seeming a little fuzzy. Hope that helps!

cupkeyk
2007-11-07, 06:38 PM
In a study of the viability of a commune where in which the population is maintained at 800, diversity in genotype can only be maintained by eugenically planned matching. 100 is way too small.

Shadowdweller
2007-11-07, 10:51 PM
Here's an illustration of the idea. Lets say we have moths that are white and moths that are black. They are in a black setting, so the white moths are more likely to die and less likely to breed. Over time, the genes conferring the white trait run a high risk of being lost.

I know that sounds a lot like natural selection, but its not. :\ I'm just having a hard time coming up with a good example.
I'm afraid that is in fact natural selection. Genetic Drift occurs from random fluctuation...cases where alleles provide neither benefit nor detriment. An example: Blue versus Brown eyes where the difference in pigment causes no differences in mortality/natality nor is one sexually selected for (sexual selection: the case of the Blah-Blah islanders considering brown eyes more beautiful in a potential mate). Simple meiotic division causes random fluctuations of blue eyed versus brown eyed populations within a closed total population, and given enough time one allele disappears entirely so that there are no longer offspring of that type.

dragonseth
2007-11-08, 01:43 AM
I'm afraid that is in fact natural selection. Genetic Drift occurs from random fluctuation...cases where alleles provide neither benefit nor detriment. An example: Blue versus Brown eyes where the difference in pigment causes no differences in mortality/natality nor is one sexually selected for (the Blah-Blah islanders consider brown eyes more beautiful in a potential mate). Simple meiotic division causes random fluctuations of blue eyed versus brown eyed populations within a closed total population, and given enough time one allele disappears entirely so that there are no longer offspring of that type.

That case of the eyes is sexual selection, by the way.

Shadowdweller
2007-11-08, 01:54 AM
That case of the eyes is sexual selection, by the way.
Incorrect. That bit inside parenthesis is assumed NOT to be the case. It is instead a counterexample; a definition of what is meant by "sexual selection". Random fluctuation of the incidence of an allele as per meiosis is NOT sexual selection.

(Original statement altered for clarity.)

AtomicKitKat
2007-11-08, 05:58 AM
Ok, so we have 50 couples(100 people total)

So we have M1 with W1, and so on till M100 with W100.

Now, let's try to make it so if they have a daughter, they marry back up the line(W100a/b and M99a/b), while a son marries down the line(M1a/b and W2a/b)

Assuming each couple produces exactly 1 child of each gender(to maintain balance of replacement, diversity, etc.) Assume also that only the males of the first generation are related, and first cousins.

So we have 100 children, with approximately 1/8 relationship to each other.

Meh, I'm not gonna do the calculations. Suffice to say that the scientists basically calculated and mating with one's first cousin is about as dangerous(genetically) as mating with a complete stranger.

The other thing that I calculated, is that it is least dangerous to mate with your paternal cousin, and most dangerous to mate with your maternal cousin(opposite sex, obviously). Why?

Your father: X(1-2)(Your grandmother's)Y1
Your father's brother: X(1-2)Y1
Your father's sister: X(1-2)X3(Your grandfather's)
You(M): X(4-5)(your mother's)Y1
You(F): X(4-5)X(1-2)
Your paternal cousin(father's brother's son): X(6-7)(Their mother's)Y1
Your paternal cousin(father's brother's daughter): X(6-7)(Their mother's)X(1-2)
Your paternal cousin(father's sister's son): X(1-2)Y2(Their father's)
Your paternal cousin(father's sister's daughter): X(1-2)X(8-9)(Their father's)

Of course, this only works with breeding out X-based recessive genes. There is still a chance that they will carry, of course, due to the mixing and matching that occurs right after the sperm and egg combine, but the odds are way better than with your mother's nephews and nieces(where you will all share Xs).

Artanis
2007-11-08, 10:58 AM
Basically the idea is that having rifted from their society the four effected families move to a provence called Voetaari and there is some issues with locals etc but basically each family moves into a different area of the country and venerates a different nature spirit that dwells there.

The familes of 100ish people mostly breed amongst themselves minimum distance of cousins however, and ocasionally other tribes send people for alliances and stuff. I want to know if its viable or if I need more principle people before I write their history.
Actually, that gives me an idea.

Say 100 isn't enough to be truly viable...without magic. You could make it so that there's the 100ish you originally intended, building up all kinds of bizarre mutations, but everybody is perfectly normal due to a blessing the nature spirit puts on each pregnancy. So an unborn child that would otherwise be born with three arms and a tail would instead be blessed in the womb and come out looking like any other baby, with the horrible genetics suppressed.

So that way, you can either have them be normal, or even play it up, making them unwilling to leave their little sanctum.

Slavakion
2007-11-08, 11:41 AM
So what does this mean? First off, families tend to have a lot of similar genes ( I should get a lot of "Duhs" for this one). So, lets say that Cousin It and Cousin What are 2nd cousins. They both got a recessive gene from their Great-Grandpa Where. Lets call this recessive gene A1. They both also have a gene A2. Neither of them actually know this since they've never gotten genetic mapping done, and the A2 gene is dominate. That is to say that they would SEEM to be like an A2A2 person - completely healthy.
Unless you're talking about epistatic genes, I do believe you're confusing genes and alleles.


What would happen after a LONG time is eventually all of those "bad" recessive genes would get bred out of a population because those with the highest fitness (not carrying the recessive gene at all) will be more likely to produce more and healthier progeny. Those who had both genes, A1A2 will probably survive to breed, but will produce less children than the A2A2 individuals.
No. If the A2 allele is dominant to the A1 allele, and A2 produces a desirable trait, then there is no phenotypic difference between A1A2 and A2A2. They have the same viability.



So its not that the family doesn't have enough genetic variation. Technically, its possible that they do and with a long enough time span they could get rid of all of their negative traits.
No. Unless you start with a stock of purebred dominant genotypes or purebred recessive genotypes, statistics will dictate that both the dominant and recessive alleles will survive in the population.

Benejeseret
2007-11-08, 12:20 PM
First off, the poster is really asking 2 questions. The first is how will the genetics work / inbreeding depression. The second and main question is about survival.

Well, we cannot even begin to answer overall survival unless we quantify survival and selection.

1. What average # of children can a female produce before death?

2. What average of children reach reproduction?

3. How severe is selection on deleterious recessive traits?


Basically, in the lab my mice can usually be bred from a single pair of mice to near infinity. Why? Because there are no predators they have all the food and resources they ever want, and we keep fights/murders down by keeping them separate. All alleles are fully fixed, with already likely over 20 generations of sib-sib breeding. They might be blind, frail, sickly...but so what, because there is next to no negative selection against these traits in this environment.


Can you make 100 people go off and survive => Yes.

As DM you control the environment, you control selection, you control 'genetic load'. If you want to be realistic, then make many such families go off. A few will survive and thrive, a few will totally die off (maybe a mudslide drowns them all on day one of hiking off), and a few will wither but hold on to a small band.

PLUS, as this is DnD make the 'genetic load' whatever you want. Maybe there us a trace of dragon/fey/elemental/demonic/monsterous ancestry that through inbreeding and positive selection becomes fixed and dominant.
A teifling has a family, two teifling descendants somewhere down the line mate and have a half-demon (template) as well as normal and teifling children. Assuming demon/human alleles are co-dominant, in theory a few teiflings mating could result in the birth of a full demon if each parent was a half-demon. If numerous populations of tieflings started to interbreed even the disperse demon alleles could all come back together in a population that selects positively for demon traits.

daggaz
2007-11-08, 12:35 PM
I study Nanotechnology, but we spend a great deal of time on genetics as well.

By 'gentically pure,' I am going to assume you mean two following things:

1. Sharing very little 'differential' DNA with the other families (the highly random bits that make humans different from eachother).
2. Containing little or no genetic disorders.

There's a catch with number two. The vast majority of people do not have any expressed genetic diseases. They do have many recessive copies of such defects. There are literally thousands of different errors that can occur in just as many genes, which can result in a big problem. Luckily we have two copies, huh? It gets more complicated due to the fact that genes are usually never simple on/off switches, and are generally influenced by any number of other genes, as well as environmental switches.

So if you just assume they are outwardly healthy?

No, you do not have enough population. It isn't really the size of the families that is the problem here, it is the number of families. Assuming each family has a random assortment of the same genes (to within a certain degree), there is only so much mixing (ie generations) that can occur before you reach the threshold, even with optimal marriages to reduce inbreeding, whereupon recessive genes begin to show up. It is pure statistics at this point. Having larger family sizes only increases the number of children, it has very little effect on number of generations past a certain point. With your numbers, you can get maybe one or two extra generations in, if you are lucky..

If you assume they are entirely free of defective genes, even recessive forms?

The population will continue to exist problem free until random genetic mutations begin introducing errors. Or the next orc invasion.

There are hundreds of RL examples of such situations btw... Pitcairn' Island, Iceland, European Aristocracy, Utila Island in Honduras, ... all of them exhibit major and obvious genetic problems at some point. It is only the influx of outsiders which has kept any of these populations going. Families which have attempted exactly as you describe (try doing some research on early european eugenicists) have ended in horrible failure.

EDIT: A lot of people are tossing numbers around like 70, 100, 500 etc. These are for genetic INDIVIDUALS. If you are talking about highly similiar family groups, you dont have genetic individuals anymore. 500 people who arent related? You can go a long ways, baby. 500 people from 4 distinct families? BIG TROUBLE.

Slavakion
2007-11-08, 12:40 PM
2. What average of children reach reproduction?
This is probably quite high, as puberty occurs in the... 12-13 area on average, I believe. If you mean what is the average of children who actually reproduce, that's another thing altogether.


3. How severe is selection on deleterious recessive traits?
Considering we're talking about a human settlement, I'd guess not very severe. A trait causing a disadvantage in pure survival would likely not be bred out (as much as possible) unless it was severely deleterious. For the sake of argument, let's say that there's a strength gene. I'm recessive for the strength gene, so I'm fairly weak. In the wild, I'd die at an early age. In a human settlement, I'd be protected and set to work on something not requiring much physical strength. Of course there's no actual strength gene, but it illustrates my point.



Basically, in the lab my mice can usually be bred from a single pair of mice to near infinity...They might be blind, frail, sickly...but so what, because there is next to no negative selection against these traits in this environment.
Exactly.

HidaTsuzua
2007-11-08, 01:30 PM
There are hundreds of RL examples of such situations btw... Pitcairn' Island, Iceland, European Aristocracy, Utila Island in Honduras, ... all of them exhibit major and obvious genetic problems at some point. It is only the influx of outsiders which has kept any of these populations going. Families which have attempted exactly as you describe (try doing some research on early european eugenicists) have ended in horrible failure.

EDIT: A lot of people are tossing numbers around like 70, 100, 500 etc. These are for genetic INDIVIDUALS. If you are talking about highly similiar family groups, you dont have genetic individuals anymore. 500 people who arent related? You can go a long ways, baby. 500 people from 4 distinct families? BIG TROUBLE.

I took it mean that there were several groups formed from a random population of 500 who then split off into their own "families" with little to no interbreeding between families.

It's true that the effective genetic population of these families will be pretty low for a long time (since it factors in the size of the previous generation's effective genetic population than numbers). However you can have such small populations. In the paper I gave mentioned 500 as the high end for Hawaiian colonization. Easter Island likely wasn't much bigger if not smaller than that.

It's a matter of scale. Yes, there will be problems. Each family will have a strong founder effect. They will have genetic diseases. But are they all going to become Int 4 slack-jaws that can barely walk? Not necessarily. Give each family some genetic diseases (you can look up South African, Amish, and jewish populations for ideas). If you're still unsure, throw in some families that "didn't make it" and were absorbed into other families. Or let some interbreeding occur and keep the families as political units (though special "exchanges" that occur every X years or a similar tradition.

Kiero
2007-11-08, 01:50 PM
What the OP is asking about is minimum viable population (http://en.wikipedia.org/wiki/Minimum_viable_population).

From a genetic standpoint "purity" is not a good thing. That way inbreeding and genetic disorders lie.

Ceres
2007-11-08, 02:35 PM
http://www.smbc-comics.com/comics/20071107.gif

Benejeseret
2007-11-08, 03:47 PM
"MVP does not take human intervention into account."

Quoted right from the wiki article linked above. Not taking human intervention into account when discussing a group of humans 'intervention-ing' themselves every minute makes any ecology equations pretty mute. Same with Hardy-Weinburg (it is theoretical and the OP's post breaks nearly every assumption needed).

Really, anthropology and examples is the best we can do. As a geneticist by training (BSc and doing PhD), I would say that deleterious (for the environment) alleles will likely become fixed at a high-than-'average' frequency....but so will positive traits depending on the selective pressures. The issue is the number or deleterious alleles and how hard the selection is.

There are soooo many assumption necessary to give an accurate prediction it is not really possible....or rather, everything is possible.

I say again, it is possible if you want it to be.


EDIT: *to Ceres* First off, funny. Second, not accurate at all in the context of the discussion...but funny. If we want to introduce artificial selection then perhaps the gods may even have a hand in the OP's world artificially selecting

NerdVicious
2007-11-08, 11:36 PM
No. If the A2 allele is dominant to the A1 allele, and A2 produces a desirable trait, then there is no phenotypic difference between A1A2 and A2A2. They have the same viability.

No. Unless you start with a stock of purebred dominant genotypes or purebred recessive genotypes, statistics will dictate that both the dominant and recessive alleles will survive in the population.

Sorry, I should have made myself a bit more clear. A1 is recessive to A2. An individual who is A1A1, in my example, would end up in some form of early death that would prevent that person getting to a breeding age.

So, if an A2A1 person produces progeny with another A2A1 person, 25% of their children, on average, will die prior to reaching an age to reproduce. This would slowly eliminate the deleterious recessive allele from the population.

I'll admit it won't FULLY take out the allele (if on a graph, there'd be an asymptote before complete fixation of the A2 allele), but the gene would be mostly bred out of the populace.

Overall, my point was the deleterious recessive alleles, when they DO pop up in the populace, tend to have a very high selection coefficient, and over time, that takes them out of the population. I gave an extreme example (selection = 1) to illustrate, but in reality, most of the time, the selection coefficent is just high, not lethal.

So, off of the fact that several individuals in the populace, per generation, will have such a low fitness, the genotype ratios cannot be maintained. Over a long period of time, those alleles with a low fitness will become increasingly rare in a population. They won't completely be bred out, but the instance of them cropping up will eventually drop off. Just would take a lot of generations.

But, whatever. I'm tired and probably repeating myself. As pointed out, this IS D&D, and I think it'd be easy to set up a scenario that would be satisfactory to anyone who might attempt to criticize the possibility of a group of 100 people living in a town. If the adventures can get there, there is always a possibiltity of new individuals entering into the population, and decreasing the effects of inbreeding. :smallsmile:

dyslexicfaser
2007-11-08, 11:52 PM
I'm hardly a geneticist, but I believe that such a small sampling could theoretically work, IF they were willing to ruthlessly weed out any children that showed genetic defects.

Too bad one smallpox blanket could probably kill the entire group.

Shadowdweller
2007-11-09, 03:34 AM
No. Unless you start with a stock of purebred dominant genotypes or purebred recessive genotypes, statistics will dictate that both the dominant and recessive alleles will survive in the population.
Not quite. Barring new sources (e.g. from mutation or influx of breeding individuals from other populations) a deleterious recessive allele such as this eventually WILL disappear. While selective pressure against a recessive allele becomes increasingly weak as incidence within a population decreases, incidence of ANY allele in an heterogeneous population fluctuates randomly. When selective pressures (repeatedly) bring down the incidence of a deleterious allele, it has a relatively high probability of disappearing. Over the long run, it WILL happen, particularly in small populations (fewer individuals = more drastic fluctuation).

Genetic disorders/defects often persist because mutation provides a steady supply of new sources in counterbalance to weak selective pressure. For many of the common disorders (not always recessive, of course), such as Down's Syndrome, new mutation greatly overshadows receipt from a carrier.

Shadowdweller
2007-11-09, 04:28 AM
EDIT: Deleted due to personal dissatisfaction

Kiero
2007-11-09, 04:30 AM
I'm hardly a geneticist, but I believe that such a small sampling could theoretically work, IF they were willing to ruthlessly weed out any children that showed genetic defects.

Too bad one smallpox blanket could probably kill the entire group.

Eliminating even more variety from an already limited gene-pool would be a very bad idea.

Dervag
2007-11-09, 04:33 AM
If the group practices infanticide to get rid of children born with double-recessive genetic problems, they may not end up with enough children to survive and breed in a dangerous environment.

Ulzgoroth
2007-11-09, 05:08 AM
Mustn't forget that there's a constant spontaneous generation rate in this. You can't have a mutant rate even asymptotic to zero.

But yeah, you can mess with usual assumptions pretty well. I can just see the genetic caste system based on ability to produce healthy offspring, with the bottom of the barrel not killed but maintained for test-crosses.

Setra
2007-11-09, 05:13 AM
Where do the half-dragon badgers come from? Where?!
I was drunk okay?! :smalltongue:

Tengu
2007-11-09, 05:53 AM
I might not know much about genetics, but sufficiently advanced geekery can replace all other knowledge!

According to TVTropes, the minimal human population needed to avoid negative effects of inbreeding is 150-160 (http://tvtropes.org/pmwiki/pmwiki.php/Main.AdamAndEvePlot). It will probably require some arranging when it comes to marriages, though.

Nnanji
2007-11-10, 12:12 AM
Sorry, I should have made myself a bit more clear. A1 is recessive to A2. An individual who is A1A1, in my example, would end up in some form of early death that would prevent that person getting to a breeding age.

So, if an A2A1 person produces progeny with another A2A1 person, 25% of their children, on average, will die prior to reaching an age to reproduce. This would slowly eliminate the deleterious recessive allele from the population.

I'll admit it won't FULLY take out the allele (if on a graph, there'd be an asymptote before complete fixation of the A2 allele), but the gene would be mostly bred out of the populace.



That's still not accurate. Granted, 25% of their children will die without breeding, however 50% of their children will be carriers of the recessive allele. Considering that only 3 of 4 children survive to breeding age, that means that 2/3 of the breeders will be carriers of the allele. Adding in the fact that other disasters (ie: natural selection occuring for some reason other than the recessive disorder such as predators) will strike evenly across the breeders, you will never create a population of all homozygous dominant. There would never be enough homozygous dominants to breed.

Shadowdweller
2007-11-10, 05:02 AM
Mustn't forget that there's a constant spontaneous generation rate in this. You can't have a mutant rate even asymptotic to zero.

While that is true, said generation rate can very easily be so small as to be effectively irrelevant. Particularly in cases where different alleles do not simply represent cases of functional versus nonfunctional proteins.


That's still not accurate. Granted, 25% of their children will die without breeding, however 50% of their children will be carriers of the recessive allele. Considering that only 3 of 4 children survive to breeding age, that means that 2/3 of the breeders will be carriers of the allele. Adding in the fact that other disasters (ie: natural selection occuring for some reason other than the recessive disorder such as predators) will strike evenly across the breeders, you will never create a population of all homozygous dominant. There would never be enough homozygous dominants to breed.
It is accurate, actually. Mortality rates remain consistently higher amongst carriers than non-carriers. At any point where there are at least two carriers capable of interbreeding, homozygous dominant individuals will continue to produce more offspring (barring disproportionate "luck", of course). As a result of which, carrier incidence will tend to decrease. Other causes of mortality are completely irrelevant unless somehow linked to the allele in question (e.g. by epistasis, side effects, or linkage on the chromosome level)...reducing the total population numbers or blanket mortality rates does not affect incidence rates.

The issue is that at low incidence rates, the mortality differences are very small...since the requirement of a carrier breeding with another carrier becomes increasingly unlikely.

preserver3
2007-11-10, 05:46 AM
As I understand it depends on the current growth of the population.

If you have 100 people who are not related, and your population size from generation to generation increases at a rate of replacement +1 per coupling, ignoring incidence of infertility and larger family size, your "group" starts suffering in 160 years or 8, 20 year generations. At 9 generations, recessive traits start showing up that cause major issues, but on top of that, you have a population of 2500+ people in the same geographical area, and the strain of maintaining that population might lead to factionalization, warfare, or even the development of more advanced communal groups like cities and countries.

I read a paper once that suggested one of the first recessive traits to show up in populations approaching their inbreeding limit was severe violent tendencies. After 3 generations of breeding with first cousins, suddenly the majority start developing violent tendencies. This paper then made the giant leap suggesting that Scottish clan wars, as well as Greek wars in the Bronze age were possibly related to a natural instinct to overcome inbreeding and sexually assault one's neighbors. He then went on to suggest the struggle between the Hatfield and McCoys was a similar struggle, but that's since been "slightly" discredited by the Von Hippel-Lindau disease crowd, who claim it was the McCoy's autosomal dominant trait combined with inbreeding that reinforced the condition throughout the family at the time of feud.

All that said, if you're trying to write a story about familial groups, start with a good base to "plagiarize" from.
Roman Soldiers Founded Chinese Town? (http://www.telegraph.co.uk/news/main.jhtml?xml=/news/2007/02/02/wroman02.xml)

146 Romans apparently took Chinese wives and settled. Though some left the region and a host of other racial drift functions were at work, this population still seems to have some European features, including partially recessive features like blond hair.

If your population can breed with the locals, it's a much more viable story.

Last note, I can't remember where it was, but there was a study that suggested that all Y chromosomes in Caucasian populations came from 13 specific ancestral fathers. They said then that it suggested stable populations were more dependent upon female populations than male populations, because mitochondrial rna suggested a much larger subset of unique mothers than the Y chromosome suggested of fathers.

Hope this helps.