Talk:Coalescent theory

I've just been adding information to Coalescence (genetics) and then came across this page. Personally I feel that the two pages should probably be merged into one article as they are essentially covering the same ground.

What existed at Coalescence (genetics) before the changes I added today was really just a very brief primer of this page, and the whole disambiguation of coalescence should point to this document.

Thoughts and discussion on the merits of doing this, and appropriate structure would be more than welcome.

Slack---line 02:53, 9 January 2007 (UTC)

I wrote most of coalescent theory, and I'm inclined to agree; there's no need for two articles on essentially the same subject. You've done some very good work so far; they shouldn't be too difficult to combine, especially your much more extensive reference list. Opabinia regalis 03:58, 9 January 2007 (UTC)

Okay, I shall wait a few days and see if there are any comments added to the discussion on coalescence (genetics) before merging the two. In the mean time I'll bulk things out there.

Any thoughts on structure for the merged document?

Suggested sections would be..

• Overview
• Theory (Probability of Fixation/Time to Coalescence/Neutrality/Extensions)
• Graphical Representation
• Applications (Phylogeneny/Disease gene mapping)
• Software (Simulation/Analysis)
• History
• References
• Links

Slack---line 00:50, 10 January 2007 (UTC)

This suggestion sounds very good. Just to note, the general convention here is to place what is effectively an "overview", referred to as the lead section, at the top of the article before the first heading. Then you'd put headings for "theory" and subsequent sections. Normally, the subject of the article is in bold text early in the lead. (ie, the format currently used by coalescence theory.)

Feel free to wait if you want, but also feel free to be bold; these pages don't get much traffic. Opabinia regalis 03:16, 10 January 2007 (UTC)

Went ahead and did the merge (bored at work as I've nothing to do), all seems to have gone smoothly. Some sections are currently blank, but I'll endeavour to gradually help bulk this out as time permits. Slack---line 05:18, 10 January 2007 (UTC)

= mtDNA

For old specimens, mitochondrial DNA is used (ancient DNA). coalescence rate will be 1/N. might be good to point that out. —Preceding unsigned comment added by 78.86.154.190 (talk) 19:39, 17 October 2007 (UTC)

Missing reference [1]

The first citation to Dawkins is missing its associated reference in the endnotes.

MidwestGeek (talk) 06:49, 17 September 2008 (UTC)

I find the sentance "Since the fixation rate is the rate of novel neutral mutation multiplied by their probability of fixation, the overall fixation rate is ... " ambiguous. Does the "fixation rate" mentioned here refer to the rate of production of new mutations that will become fixed at some future time, as opposed to the rate of fixation events/number of mutations fixed per generation? —Preceding unsigned comment added by 84.13.45.218 (talk) 21:20, 7 May 2009 (UTC) The fixation rate is the proportion of mutations that reach fixation.

But in my opinion this paragraph has nothing to do with the coalescent, but is a classical result of standard, forward-time population genetics. Coalescent theory is a "backward in time approach", i.e. I look at my sample and wonder which processes led to its allele frequency distribution, In this context fixation as something that might happen in the future is not relevant at all. The classical fixation index FST has been redefined by some authors as "probabilities of coalescence event before a lineage leaves some deme", but I doubt this fits in the scope of this article. Jixani (talk) 21:05, 9 June 2009 (UTC)

I moved the paragraph about Fixation time to Fixation (population genetics). --Ettrig (talk) 07:06, 10 June 2009 (UTC)

Hey folks

I was reading this article and it would appear that coalescence theory is attributed to John Kingman, and while that may be true for coalescence theory, someone completely overlooked the originator of the formula not even a single reference was provided.

Process leading to quasi-fixation of genes in natural populations due to random fluctuation of selection intensities. Motoo Kimura. Genetics 39:280-295 (1952) PMID: 17247483

Variance of the distribution in the t-th generation is from (22), (23), and (26)

Vt = pq - pq * e ^(-1/2N)

and namely the variance approaches its limiting value of pq at the rate of 1/2N per generation. As has been demonstrated above the process of change due to random fluctuation of selection intensite is quite diffrent from that due to the random sampling of gametes. Therefore comparison of their effects must be made from various angles as Wright (1948( did in analyyzing the data of medionigra gene in Panaxia.

Thus if we consider the process of change which will be realixed after a sufficient number of generations, the rate of quasi-fixation, Vs/8 maybe compared with the rate of fixation due to random sampling, 1 / 2N, fo the same purpose. Suppose Vs is known to be 0.0483 as in figure 1, the equivilent N is calculated to be about 83 by using a equivalence relation;

1/2N = Vs/8

Kimura effectively lays out fixation parameters in Figure 1 parameters indicating that at a starting frequency of 0.5 that the probability versus frequency distribution first flattens, becomes bimodal and then until t = 500 the probability migrates toward a o probability of a 0.5 freuquency and .50 probability of a 0 frequency or 1 frequency.

p.294

Confusing phrase

I found the phrase "A useful analysis based on coalescence theory seeks to predict the amount of time elapsed between the introduction of a mutation and a particular allele or gene distribution in a population." a bit hard to understand at a first reading, for someone outsider or newcomer to the matter, as the time in wich the said distribution exists is not clear.

What about inserting "A useful analysis based on coalescence theory seeks to predict the amount of time elapsed between the introduction of a mutation and the arising of a particular allele or gene distribution in a population." to better clarify the 'present time' idea?

Lineage sorting

Redirects here, but the term appears nowhere in the article. While this is OK for the time being, lineage sorting has actually little to do with coalescent theory except being one of the reasons why coalescent theory has practical applications (if it weren't for lineage sorting, the genotypes compared in coalescent theory could not be considered representative of populations). It should eventually get its own article, or perhaps be treated under population genetics#Genetic_structure. Dysmorodrepanis (talk) 22:27, 18 August 2011 (UTC)

References

I find that the coalescent theory does not sound enough as a relevant stochastic theory, there are new approaches since Pitman and Möhle, Bertoin contributes in 2006 to the theory (this was a bit hidden because of fragmentation theory) but his method simplifies consequently the definitions and for a mathematician could also be a reference. Coalescent theory became a totally relevant domain in probability, and the monography from Nathanael Berestycki could be more well presented. Waiting for someone to introduce the current mathematical goals in coalescent theory. — Preceding unsigned comment added by 78.192.27.35 (talk) 00:01, 12 May 2013 (UTC)

---

>do it!