On the other hand, Rana seems to be evaluating this evidence under the mistaken impression that "It is possible to estimate the effective population size of any ancestral group from genetic diversity of present-day populations if the mutation rate is known." That's not how LD works. It's actually a lack of mutation that indicates LD. My fellow creationists (old earthers and young earthers alike), if we're going to formulate a response to LD-based population size estimates, we first need to understand how the LD evidence works.
Now, this is just a small piece, but it seems to me that Rana is saying that, to estimate the size of the population using these methods, we have to know the rate of mutation. Wood's response is, no, it's not the presence of mutations that indicates LD, but the lack of mutations.
But (at least given the quick quote) Rana doesn't seem to be saying otherwise. Whether you say "there are too many mutations" or "there are too few mutations", it seems like either estimation is going to involve known the rate of mutations and determining if what you see matches up with what you should expect to see.
It's a little like saying, "Knowing the rusting process for metal will let you tell how old a given artifact is" and someone replying, "But it's not the presence of rust that tells us the artifact is young, but the lack of rust."
Not to mention, the entire tone of the article is just odd. It's supposed to be an evaluation of Venema's arguments and evidence. Instead, it's 95% criticizing RTB, 5% frankly kissing Venema's ass.
Though I wouldn't put it in those words, the blog article wasn't what I was expecting either. :)
Todd Wood says he doesn't know how to answer Venema, but the study should include other "species" too. And then he moves on to RTB. I don't think he was ever intending much of an evaluation of Venema's arguments.
1. It sounds to me Wood would say he'd agree with Rana's conclusion but not with how Rana arrived at his conclusion.
2. With regard to linkage disequilibrium (LD). Genes or alleles are said to be in LD when the alleles at two different locations (loci) occur more frequently in combination than would be predicted based on random assortment i.e. chance. This can be used for purposes such as mapping the genetic basis of a particular disease or dating genetic events. Or so the argument goes.
For example, let's take an extremely simple case. Say the frequency of an allele in a population is 0.15 or 15%, while the frequency of another allele is 0.20 or 20%. Based on random assortment or chance, one would predict the two alleles would be found together with a frequency of 0.15 x 0.20 = 0.03 or 3% of the time. However, say in reality the two alleles are found together 15% of the time in the population. So, since 15% is 5.0 times greater than 3%, the two alleles are found together 5.0 times more frequently than expected or predicted by their individual allele frequencies i.e. random assortment or chance. Thus, this disequilibrium suggests linkage of the two alleles on a specific locus or loci which is on, say, chromosome 7.
Now, with the passage of enough time, or rather with the passage of enough successive reproducing generations, the frequency of one allele in a population should be near identical to the frequency of the other allele in individuals with or without the other allele. In other words, they have come to equilibrium.
I take it this is what Wood is getting at.
3. With regard to mutation rates. It's hard to figure out mutation rates for many reasons. For one thing, many mutations are silent. For another, mutation rates can vary in different genes. For still another, we have to differentiate between germline vs. somatic mutations (for starters). For yet another, environmental factors can affect mutation rates (e.g. check out this article). Anyway, there are more reasons, but I'll stop for now.
Something bothers me here.
ReplyDeleteOn the other hand, Rana seems to be evaluating this evidence under the mistaken impression that "It is possible to estimate the effective population size of any ancestral group from genetic diversity of present-day populations if the mutation rate is known." That's not how LD works. It's actually a lack of mutation that indicates LD. My fellow creationists (old earthers and young earthers alike), if we're going to formulate a response to LD-based population size estimates, we first need to understand how the LD evidence works.
Now, this is just a small piece, but it seems to me that Rana is saying that, to estimate the size of the population using these methods, we have to know the rate of mutation. Wood's response is, no, it's not the presence of mutations that indicates LD, but the lack of mutations.
But (at least given the quick quote) Rana doesn't seem to be saying otherwise. Whether you say "there are too many mutations" or "there are too few mutations", it seems like either estimation is going to involve known the rate of mutations and determining if what you see matches up with what you should expect to see.
It's a little like saying, "Knowing the rusting process for metal will let you tell how old a given artifact is" and someone replying, "But it's not the presence of rust that tells us the artifact is young, but the lack of rust."
Am I off-base about this?
Not to mention, the entire tone of the article is just odd. It's supposed to be an evaluation of Venema's arguments and evidence. Instead, it's 95% criticizing RTB, 5% frankly kissing Venema's ass.
ReplyDeleteCrude,
ReplyDeleteThough I wouldn't put it in those words, the blog article wasn't what I was expecting either. :)
Todd Wood says he doesn't know how to answer Venema, but the study should include other "species" too. And then he moves on to RTB. I don't think he was ever intending much of an evaluation of Venema's arguments.
1. It sounds to me Wood would say he'd agree with Rana's conclusion but not with how Rana arrived at his conclusion.
ReplyDelete2. With regard to linkage disequilibrium (LD). Genes or alleles are said to be in LD when the alleles at two different locations (loci) occur more frequently in combination than would be predicted based on random assortment i.e. chance. This can be used for purposes such as mapping the genetic basis of a particular disease or dating genetic events. Or so the argument goes.
For example, let's take an extremely simple case. Say the frequency of an allele in a population is 0.15 or 15%, while the frequency of another allele is 0.20 or 20%. Based on random assortment or chance, one would predict the two alleles would be found together with a frequency of 0.15 x 0.20 = 0.03 or 3% of the time. However, say in reality the two alleles are found together 15% of the time in the population. So, since 15% is 5.0 times greater than 3%, the two alleles are found together 5.0 times more frequently than expected or predicted by their individual allele frequencies i.e. random assortment or chance. Thus, this disequilibrium suggests linkage of the two alleles on a specific locus or loci which is on, say, chromosome 7.
Now, with the passage of enough time, or rather with the passage of enough successive reproducing generations, the frequency of one allele in a population should be near identical to the frequency of the other allele in individuals with or without the other allele. In other words, they have come to equilibrium.
I take it this is what Wood is getting at.
3. With regard to mutation rates. It's hard to figure out mutation rates for many reasons. For one thing, many mutations are silent. For another, mutation rates can vary in different genes. For still another, we have to differentiate between germline vs. somatic mutations (for starters). For yet another, environmental factors can affect mutation rates (e.g. check out this article). Anyway, there are more reasons, but I'll stop for now.
Maybe a picture is worth a thousand words (source)?
ReplyDelete