Talk:Quantitative trait locus

From Wikipedia, the free encyclopedia

	To-do list for Quantitative trait locus:	edit  · history  · watch  · refresh
describe the biology behind the analysis, how the analysis is performed, and the reasons why it works. see Josh's list below

I think that this sentence doesn't fit it, nothing has been said about growth hormone before. "In this way, we might find a gene involved in the production of growth hormone." can the author check it please?

Contents 1 History and confusion 2 Merge Quantitative trait gene into Quantitative trait locus 3 Suggested merges 4 Rename article

[edit] History and confusion

The sentence "The QTL techniques were developed in the late 1980s and can be performed on inbred strains of any species" is very confused. In fact, this entire article seems to think that interval mapping is the same as QTL analysis, which is not the case (although this is the common usage today).

First off, "QTL techniques" were around well before the late 1980s. The first example I know of QTL analysis is from Thoday in the 60s and 70s (see Genetics and Analysis of Quantitative Traits by Lynch and Walsh) -- and there were even chromosomal association studies before that in Drosophila [actually, at 3am I discovered "The association of size differences with seed-coat pattern and pigmentation in Phaseolus vulgaris", Sax K, Genetics 140:201-206]. Between Thoday and Lander and Botstein (see later), QTL analysis got a lot more sophisticated, with more sophisticated marker genotyping methods. The problem with these methods is that while they could detect a QTL, they couldn't tell you where it was (as far as I know).

In the late 1980s Lander and Botstein proposed a new way to use dense markers (in their case, RFLPs, or restriction fragment length polymorphisms). In their method, pairs of flanking markers are used to estimate the likelihood that a QTL exists at every position in the genome. This is how the LOD maps are generated (the pretty images in this article).

So, some advice I have for this article:

• Add in some history of QTL analysis that includes Thoday and the many others after him and before Lander and Botstein.
• Have a better explanation of the statistics behind interval/LOD mapping (what this article calls "QTL mapping"). I.e., an LOD score is the log of the ratio between the maximum likelihood that the data fit a null hypothesis and the maximum likelihood that the data fit a QTL hypothesis, given the observed phenotypes and marker genotypes (see Lander and Botstein, "Mapping Mendelian Factors Underlying Quantitative Traits Using RFLP Linkage Maps", 1989).
• Fix the error that QTL mapping can only be performed in inbred strains. If this were the case we wouldn't have QTL maps for human traits (like osteoporosis, the example in the article). What's required is linkage disequilibrium between a QTL and a marker (since, if the traits were in Hardy-Weinberg, there would be no association between a marker genotype and a QTL genotype), which is maximized in inbred strains but can also be achieved in the wild, by examining drift-caused disequilibrium in families or expanding populations.

-- Josh 06:57, 7 March 2007 (UTC)

It is stated that QTL mapping is only done in inbred lines, but QTL mapping can be done in outbred lines as well, as long as the markers that are used for mapping are consistent within each line but differ between the two lines of interest. It can even be done between closely-related species (to identify QTL for species differences) with the caveat that the species need to still be able to mate and produce viable and fertile offspring.

[edit] Merge Quantitative trait gene into Quantitative trait locus

Both major textbooks, Falconer & Mackay, 4th ed., and Lynch & Walsh, do not use the term "quantitative trait gene". Locus is a general term that can refer to a gene or any other genomic feature. "Fine mapping" can equally well apply to QTL mapping.

In my MSc course on quantitative genetics (possibly the only full course on the subject) at Edinburgh Uni (where much of this theory was developed), I did not come across the term. It does seem to be used by at least one academic, namely any one or several of Z-B. Zeng, C. J. Basten, R. Doerge, B. Yandell and I. Hoeschele [1]. - Samsara _contrib ^talk 12:53, 11 February 2006 (UTC)

I would suggest to delete it entierly, and just explain the term when needed at spots where it is used. If as all, it is more used in molecular genetics to indicates a specific gene that contributes to a quantitative trait determined by many genes. If not delete, to quantitative genetics --KimvdLinde 15:45, 11 February 2006 (UTC)

I've made that change. I've also suggested merging polygenic inheritance into quantitative trait locus - any thoughts on that? - Samsara _contrib ^talk 08:54, 15 February 2006 (UTC)

I do not agree with merging polygenic inheritance with quantitative trait locus as the first one deserves its own page with a lot more text than it has now - though, one will have to write that. 08:26 (CET) 2006-10-27

I also believe polygenic inheritance deserves its own article as it is fundamentally different from quantitative trait loci. I have added a brief overview to polygenic inheritance and have also redirected polygenic to that page. Raetzsch 05:50, 15 November 2006 (UTC)

[edit] Suggested merges

I am suggesting to merge both candidate gene and polygenic inheritance into this article, since

1. neither article goes beyond the remit of a dictionary entry
2. the candidate gene approach is an integral and exclusive part to trait mapping exercises, of which QTL mapping is the more general and powerful version
3. it is intellectually more elegant and leads to a better understanding to accommodate polygenic inheritance within a quantitative/discrete/threshold trait framework (oh dear, this one may be controversial...)

Your thoughts? - Samsara _contrib ^talk 18:36, 15 February 2006 (UTC)

Candidate gene approach can be the same as QTL, but can leave QTL's out of the picture as well. So, leave than one out. polygenic inheritance and Quantitatve trait gene should come here. --KimvdLinde 18:43, 15 February 2006 (UTC)

Several different methods, including chipa assays, LD mapping, QTL, and standard molecular population genetics can all use a candidate gene approach. I don not think it is in any way synonymous with QTL mapping. Understanding polygenic inheritance is clearly important for an appreciation of QTL mapping, but it is more general and shouuld not be merged here. Rather, polygenic inheritance should be merged with the article on Quantitative Genetics.

[edit] Rename article

After several merges, this article is no longer exclusively about QTLs. Rather, it includes all aspects of quantitative inheritance, including QTLs. Thus, I believe a better name for this article would be "Quantitative inheritance" or "Quantitative traits." QTL could then redirect to here. - Raetzsch 17:47, 10 December 2006 (UTC)