How does selective breeding work?
Post Reply
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
25-07-2012, 06:18 PM
How does selective breeding work?
Assume we want to create a new dog breed with very short snouts. We select for short-snouted dogs from a given population and isolate them genetically to allow them to inter-breed for many generations. However, why do the descendants have even shorter snouts than the ancestors? What causes the snouts to shorten even more over generations from short snouts to extra short snouts, if we never had the gene or the allele for the extra short snout?
Find all posts by this user
Like Post Quote this message in a reply
25-07-2012, 07:29 PM
RE: How does selective breeding work?
You just keep selecting shorter and shorter snouts. Occasionally you breed in a long nose so you maintain a vigorous strain, too much selection creates weak lines in the long run. It's a bad example though because short snouts bring health issues.

A good example is the German Shepherd as bred in Germany. They have been bred for hundreds of years for a number of traits: Balanced structure, intelligence, desire to learn, superior scenting, good disposition and fearlessness.

So what they have done is they designed tests or "trials" for the dogs. Only the dogs who pass the physical exam, the tracking (scenting) test, the obedience test (learning ability and willingness), the character test (friendly around people and other dogs) and the protection test (fearlessness and courage) are allowed to be bred. It takes a minimum of two years of training to be able to pass the test, and serious training is started with dogs that are one year old. Hip and elbow health is also tested with xrays at one year old. So, a dog has to be around 3 years old to see if they will be approved for breeding. All this gets meticulously recorded and goes into the pedigree, which is long and very precise.

Now that is a complicated selection because it demands so many traits. But it does end up so only the best can breed.

Another type of selective breeding that is done a lot with various dog breeds is for size. Here you run into issues of viability and defects. You are bound to run into such issues whenever you breed for only one trait and allow otherwise inferior animals to procreate.

Selective breeding is also used with plants to develop stronger or differently colored specimen etc.

Whenever you do selective breeding you have to make sure to "out cross" periodically. That means brining a different, more robust type into the program, or at least use animals with different ancestry who do not show the drawbacks you are running into.

There is also line breeding, and there you also have to out cross frequently.

Se here also: http://rationalwiki.org/wiki/Artificial_selection

[Image: dobie.png]

Science is the process we've designed to be responsible for generating our best guess as to what the fuck is going on. Girly Man
Find all posts by this user
Like Post Quote this message in a reply
25-07-2012, 07:31 PM
RE: How does selective breeding work?
Major morphological changes like a shorter snout would take some time to acquire. You would need several different groups of dogs to mate because inbreeding would lead to possibly fatal physical deformities. Independent random mutations in dogs in several of the groups might cause them to be born with slightly shorter snouts. Mating these two would of course produce a shorter snout. The snouts would continue to get shorter as you mated dogs in this way. Again, you would need several groups of unrelated dogs to do this.

Dogs actually have shorter muzzles than their wolf ancestors. A Russian experiment from the mid-20th century explains why this is. Scientists took a population of Russian Silver Foxes and mated the ones that were more docile. They continued to do this with the resulting offspring for generation after generation. Each generation naturally became tamer than the last. But by the 8th generation, they found that the offspring had shorter muzzles, floppy ears, fur patterns not known to the species, wagging tails, and a need to please humans. The foxes were pretty much turned into dogs. This transformation is tied to the level of adrenaline in their system. The offspring had far less of it coursing through their bodies, making them more docile. This affected their physical forms as well. It caused the foxes to regain juvenile traits that their pups usually have. This is known as Neoteny.
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 1 user Likes ghostexorcist's post
25-07-2012, 07:43 PM
RE: How does selective breeding work?
Hey, evgenia.

I never post in this forum. I just came to post link to a video in another thread. But I'll take a shot. If you want a follow up, PM me because I won't check otherwise.

It's about representation.

When you're breeding, you're working with a closed gene pool. For example, you don't want chihuahuas humping your dobermans if you want to breed dobermans. So step one is exclude all of the obvious suspects from your breeding stock. Essentially, you're eliminating gene flow and creating an artificial genetic drift.

Now, if you want to move from big snouts to small snouts, you have to increase the representation of the small snout "gene" (long day, I'm using short hand). So you set some threshold. Dogs born with 20cm snouts and above are no longer allowed to breed. Then you'll have some with 19cm, 15cm, 17cm, 10cm. Then you set a new threshold. No dog >18cm. Basically, you're reducing the representation of what you don't want, and increasing the representation of the variation you do want. If the short snout gene has a 10% representation, then you aren't gonna have a lot of dogs with short snouts. But if you increase its representation to, 80, or 90%, most will have it. Once you get to that point, you just set your final threshold for the breed. Dogs with snouts between 11cm and 9cm are allowed to breed. No one else. You're pretty much set after that.

Peace and Love and Empathy,

Matt
Find all posts by this user
Like Post Quote this message in a reply
26-07-2012, 06:48 AM
RE: How does selective breeding work?
(25-07-2012 07:31 PM)ghostexorcist Wrote:  Major morphological changes like a shorter snout would take some time to acquire. You would need several different groups of dogs to mate because inbreeding would lead to possibly fatal physical deformities. Independent random mutations in dogs in several of the groups might cause them to be born with slightly shorter snouts. Mating these two would of course produce a shorter snout. The snouts would continue to get shorter as you mated dogs in this way. Again, you would need several groups of unrelated dogs to do this.

Dogs actually have shorter muzzles than their wolf ancestors. A Russian experiment from the mid-20th century explains why this is. Scientists took a population of Russian Silver Foxes and mated the ones that were more docile. They continued to do this with the resulting offspring for generation after generation. Each generation naturally became tamer than the last. But by the 8th generation, they found that the offspring had shorter muzzles, floppy ears, fur patterns not known to the species, wagging tails, and a need to please humans. The foxes were pretty much turned into dogs. This transformation is tied to the level of adrenaline in their system. The offspring had far less of it coursing through their bodies, making them more docile. This affected their physical forms as well. It caused the foxes to regain juvenile traits that their pups usually have. This is known as Neoteny.

Could you please explain why the level of adrenaline (epinephrine) decreased over generations? So far as I understand, the ancestors would have genes coding for so much epinephrine in their bloodstream with some variation within that population. Let's say gene A is too much epinephrine, gene B is average epinephrine, gene C is too little epinephrine. Now, we take the offspring with less epinephrine in their blood (gene C) and breed them. Their offspring in turn seem to exhibit genes B (average), C (too little), and D (way too little). My question is, where did that gene D come from, if we only worked with A, B, and C so far?
Find all posts by this user
Like Post Quote this message in a reply
26-07-2012, 10:04 AM
RE: How does selective breeding work?
Hey, Evgenia.

I got your PM.

The question is, why does the shorter snout show up at all?

Say you begin with a species that averages a 25cm snout. The important thing to remember is that this is an average. There is no gene for the perfect 25cm snout. There's always, thanks to diploidy; you get one allele from mom and one from dad (then you deal with dominant/recessive and co-dominance); and thanks to mutation, a range of heights. So what you're trying to do is reduce that range by breeding from the bottom end of it.

So, the range progression might look like:
1 - 28-22cm
2 - 26-20cm
3 - 24-18cm

And so forth. When you get to your ideal range, there's still going to be a range, it'll just be in the neighbourhood you want.

The other thing you can do is breed in other breeds. For example, the doberman is basically a cross between a rotweiller and a greyhound. You could breed rotweillers into skinnyness, or just skip the process and inject skinnyness by mixing in greyhounds.

But basically, the answer is, the gene for height or length or whatever is likely the exact same gene for shortness. It's just a matter of how long it's turned on for.

For example, there is only one gene for human arms, dog legs, dolphin fins, bird wings and lizard legs. It's the same gene: the tetrapod gene if you will. But the end product is so different because of how long it's turned on for in utero.

So the gene is always there. You're just playing with the recipe.

PM me if you want more.

Peace and Love and Empathy,

Matt
Find all posts by this user
Like Post Quote this message in a reply
26-07-2012, 07:20 PM
RE: How does selective breeding work?
Hey, Evgenia.

Don't feel bad about PMing me. I'm good. Ask away.

I should clarify that I am not a geneticist nor am I a biologist. I am however a Darwinist. So my understanding is of evolution and selection. There are gaps in my practical knowledge that the others seem to be filling in.

If you don't mind, I'll just paste your question from the PM. If you have an issue, let me know and I'll delete it immediately.

Quote:Given that the allele combinations for the 18 cm snouts are very rare but are still in the gene pool, is that still possible that an individual with a 30 cm snout accidentally gives birth to an individual with an 18 cm snout (mathematically it is very VERY unlikely, but still possible)?

My understanding is that there is no gene or allele or anything that says, "This animal will have an 18cm snout." It doesn't work that way. If only just because of phenotypic expression.

If you have a mom and a dad and they're both 6'6, you might expect to have a child in that area.
If you have a mom and a dad and they're both 5'0, you might expect to have a child in that area.

What you will not expect is for 6'6 parents to give birth to a 5' child and vice versa.

But that doesn't mean they'll give birth to a 6'6 child. I met a man recently who was 6'2, his wife was 5'7. Their daughter was 5'9 and their son was 6'5. So we have to remember that there's a range and some randomness involved.

But if I was breeding shorter humans, I sure as shit wouldn't let their 6'5 son stud.

But there's limits. I don't know exactly where they lie but if Shaquile O'Neil had a 4' daughter, we'd all be scratching our head.

The other thing we need to remember is that the gene pool is being deliberately narrowed. There is a deliberate attempt to drastically reduce diversity.

At any rate, I don't imagine it's impossible for a 30cm snout dog to give birth to a 18cm snout dog. But if an 18cm dog gave birth to a 30cm dog, the important thing is that the breeders would not let that 30cm snout dog sire offspring.

I may have glossed over the first part of the question. I'm hoping not, but if I did, point it out. My response to the first part of the question is that the attempt is to increase the representation of those genes that lead to shorter snouts and reduce the frequency of those genes that lead to longer snouts.

Peace and Love and Empathy,

Matt
Find all posts by this user
Like Post Quote this message in a reply
26-07-2012, 08:26 PM
RE: How does selective breeding work?
Excellent explanations all! You also have to factor in the recessive genes that will show up as a dominant trait if both parents carry the gene. So even the best efforts at sticking with say, one size, won't always result in a litter of only that size.

Eye color is a great example for people and could be searched easily.

I have two Cavalier King Charles Spaniels and if you look at the history of the breed, they were specifically bred to have smaller muzzles than most of the spaniel breeds. There is a pretty big range on 'acceptable' size of the dogs overall...both mine tend to be on the small end of the range.

I was fortunate at an early age to watch genetics and selection at work. My dad was a veterinarian. My sister wanted pets and she and dad decided on rabbits. Dad took it as an experiment. He would buy rabbits of differing colors and hair length and breed then to see what the bunnies would look like. Of course, in the beginning, it was really random since we didn't know the parentage of the rabbits he bought. After a while, when we were breeding rabbits that we new some of the lineage, he was able to make very educated estimates on what the color and hair length would be in the new litters. Since bunnies breed like...well bunnies, it wasn't long before he could go back to his records and we could just about pick what color and hair length we wanted and know which to breed to get that result. Now and then the odd one would show up proving the effects of recessive genes. It was interesting to observe firsthand.

I'm not anti-social. I'm pro-solitude. Sleepy
Find all posts by this user
Like Post Quote this message in a reply
26-07-2012, 10:01 PM
RE: How does selective breeding work?
(26-07-2012 10:04 AM)Ghost Wrote:  Hey, Evgenia.

I got your PM.

The question is, why does the shorter snout show up at all?

Say you begin with a species that averages a 25cm snout. The important thing to remember is that this is an average. There is no gene for the perfect 25cm snout. There's always, thanks to diploidy; you get one allele from mom and one from dad (then you deal with dominant/recessive and co-dominance); and thanks to mutation, a range of heights. So what you're trying to do is reduce that range by breeding from the bottom end of it.

So, the range progression might look like:
1 - 28-22cm
2 - 26-20cm
3 - 24-18cm

And so forth. When you get to your ideal range, there's still going to be a range, it'll just be in the neighbourhood you want.

The other thing you can do is breed in other breeds. For example, the doberman is basically a cross between a rotweiller and a greyhound. You could breed rotweillers into skinnyness, or just skip the process and inject skinnyness by mixing in greyhounds.

But basically, the answer is, the gene for height or length or whatever is likely the exact same gene for shortness. It's just a matter of how long it's turned on for.

For example, there is only one gene for human arms, dog legs, dolphin fins, bird wings and lizard legs. It's the same gene: the tetrapod gene if you will. But the end product is so different because of how long it's turned on for in utero.

So the gene is always there. You're just playing with the recipe.

PM me if you want more.

Peace and Love and Empathy,

Matt

Something to add on to Ghost's post. Majority of our physical traits (height, weight, skin colour, etc) are not controlled by one gene, but many genes. The effects of these genes are additive. This gives a whole range of phenotypes due to the possible permutations of the alleles one might have received from the parents. This is known as continuous variation. Furthermore, such physical traits are also affected by the environment, starting in-utero, up till death. The range of phenotypes when plotted on a graph, will give you a bell curve. Here's one good example:
[Image: f11-17_height_in_human__c.jpg]

Hope this helps Smile

Welcome to science. You're gonna like it here - Phil Plait

Have you ever tried taking a comfort blanket away from a small child? - DLJ
Find all posts by this user
Like Post Quote this message in a reply
Post Reply
Forum Jump: