Thoughts and questions about measuring temporal changes in population genetic structure, genetic drift and mutation
Eitan Reem
Friday, 24 July 2009 20:17 UTC
I am working on temporal changes in population genetic structure of a marine
organism that resides in a relatively isolated habitat which means that geneflow is hardly a factor.
I have samples that were collected deliberately and consecutively over a period of 13 years.All samples were collected once in a year on the same month from the same place by the same person, using the same method of sampling. I use 5 polymorphic microsatellite loci to study the genetic structure. Average generation time of the organism in natural conditions is about 2 months -which means that there is no overlapping of generations and each sample is about 5 generations younger than the previous one.
POINT/QUESTION 1: I intend to use D(Jost) as a differentiation measure by considering each sampling year as a different population. Any comments or objections?
POINT/QUESTION 2:About using Gst or one of it’s relatives (Fst or Rst)to measure the relative balance of genetic drift to mutation rate. Usually Gst measures the relative balance of genetic drift to geneflow. As you see I substitute between these 2 evolutionary forces. I do it for the following reasons:1) “Geneflow and mutation behave in an analogous manner with respect to genetic variation within a local deme”(POPULATION GENETICS AND MICROEVOLUTIONARY THEORY by A.R. Templeton 2006 page 174)
2)Following the above citation I think that we can look at both mutation and geneflow as active sources for new alleles. While geneflow is an external source and mutation is an internal source.
If my assumptions are correct I can indeed use Gst or its relatives the relative balance of mutation to drift.
Any comments or objections?
Thanks in advance
Eitan
-
Replies
-
You’ve got a dynamic system, so why not model it as such? You could extend the methods I used a couple of years ago. Adding migration/mutation is straightforward.
I can provide the BUGS code, and help you change it for your data, but I’m on holiday at the moment, and will probably be busy until September, so I can’t give it proper attention for some time.
-
Thanks Bob. I tried to find about the method you mention in your reply by pressing the link. It transferd me to the Meta press site but there I lost my way. I submitted your name in the box, and the rusult was a long list of articles and book chapters but your name did not appear. Can you help?
Eitan -
Oops. This is the full reference:
O’Hara, R.B., 2005. Comparing the effects of genetic drift and fluctuating selection on genotype frequency changes in the scarlet tiger moth. Proc. R. Soc. Lond. B, 272: 211-217.
It looks like the whole of January has disappeared on the Proc. R. Soc. B. web page: it starts in february on p219. Hm. Someone hates me…
Luckily, there is a pdf in hawaii. Enjoy!
-
Eitan, regarding your Point 1, I think D used between the same population at different times is an excellent way to quantify the change in allele frequencies. I have been experimenting with this kind of application as well.
About your Point 2, yes, I agree that if migration is low and mutation rate is high, immigration will act much like mutation in maintaining within-group diversity.
If you are trying to quantify the balance between migration and mutation, you could try calculating absolute migration via GST and then using the method I posted earlier in this forum to find the absolute number of mutations arising per generation.
Lou -
Thanks Lou
The temporal change in allel frequency among the samples is continuous but not consistent and I can even say stochastic. For example an allele can appear in high frequency (>0.3) in one year and a year later to appear in a frequency of 0.02 or less, and vice versa. This is true with 4 out of the 5 loci I use as microsatellites.
I am trying to find possible reasons to that phenomenon,so that quantyfing of the balance between geneflow and mutation is probably an esential step.
If you have some more (good) advices I shall be glad to know about.
By the way: I want to ask another question but it is solely about D so I will post it to the JostD forum.
Eitan -
Mutation won’t cause those changes: it’s hard enough to see any mutations in microsats.
Do you have, or can you get, an estimate of Ne? With that, you can get an idea about drift.
Do you know anything about the populations individuals might be migrating from?
Also, what are your sample sizes? a lot of what you’re describing might just be due to small samples in each year.
Oh, and is your organism sexual or asexual?
-
Hi Bob!
Thanks for your cooperation. Regarding your questions: I am answering from the last to the first: Yes my organism is sexual. Sample sizes are around 30(1 sample is 15 but 3 samples are 43). I was also thinking that such fluctuations are due to small sample sizes and/or sampling error. However a study done by a colleague of mine on the same organism from another part of the world showed similar picture of temporal fluctuations, and his sample sizes were between 43-54. The organism is a marine invertebrate, possible migrants to “my” population can come from everywhere in the Pacific Ocean but most probably from other neighbour parts of the coast of California. This organism is not an active swimmer it is a sessile one. Its colonies adhere to floating objects or slow moving invertebrates with hard shells(oysters,crabs etc).Catching of crabs or oysters and transporting them from one site to another is known to be a common way of its distribution.
Regarding Ne I will try to estimate it. Anyway the real population in the sampling site is very big.
You write: “mutations won’t cause such changes” My inclination is to agree with you, however these microsatellite loci are highly polymorphic (between 11-64 alleles). There are more then 6 studies that support this . In 3 out of the 5 microsatellites I use, the repeat units of core sequences are of 2 nucleotides. This means that mutation rate can be at the 1/100 range. So I would not reject the possibility of mutation so quickly. On the other hand there is probably some rate of geneflow as I described above.
Actually my goal is to asses what are the forces that shape the population structure I found. Beside genetic forces I am also considering environmental changes (water temperature and salinity) and their posible impact on the genetic structure.
Any helpfull comments?
Thanks in advance
Eitan -
Hi Bob
There is something important I forgot to say:“my” organism is highly inbred i.e heterozygotes defficiecy and positive inbreeding coefficients – this was proved in many previous genetic studies as well as from direct observations.
My results also confirm it.
Eitan
-
