Non-mendelian genetics/ non-darwinian evolution

I just saw an interesting article about horizontal gene transfer (HGT) on the new scientist website. The article describes a study in which soil bacteria exposed to artificial lightning were found to have taken up exogenous DNA from their environment. The suggestion from the article was that a similar sort of naturally ocurring electroporation could partly account for the apparent high frequency of HGT.This, as well as Salty's recent drive-by, put me to wondering to what extent new discoveries of various non-mendelian genetic systems or exceptions to the generally recieved Darwinian inheritance are being fully integrated into peoples general understanding of evolution.I know that one of the bugbears for many of us is that creationists frequently represent all proponents of evolution as being strict literal adherents to Darwin's writings as if they were some sort of religious doctrine, as some indeed claim they are, but while the inertia in the field of evolutionary biology is clearly not that bad is there not some truth that as more and more exceptions to the various basic foundations of the modern synthesis accrue not enough is being done to give the whole field the sort of radical overhaul it may soon need?Perhaps this is being done outside of the scientific literature, I read very little popular science texts now but more and more people seem to prefer to produce new syntheses of current thinking in the form of saleable popular books rather than as peer reviewed literature, a trend which has allowed the ID movement to make a lot of headway with a surface appearance of scholarly parity.Are there new synthses of evolutionary thinking out there which are accomodating the more recent findings of epigenetics, and other unusual forms of inheritance/gene transfer?TTFN,WK

Thread moved here from the Proposed New Topics forum.

I just got a book out on electrolytic REACTIONS and as I read it I will try to THINK more about current densities modified (this is what affects rate changes) not by artifical man-made OVERVOLATGES in the lab but as you found INLIT by lightN. It would be great to be able to take my lab based confidence in applying Faraday (on what is in this case I will apply to this information but 'resistance") BACK into the Field of Artifical Natural Selection.

Hey WK,Definitely an interesting topic, I would love to see examples violating the modern synthesis.However, I have seen "non-Mendelian inheritance" used and misused in a number of ways. Simply because something is labeled "non-Mendelian" does not mean it actually violates Mendelian inheritance at the genetic level.One way I feel it is severely misused in the literature is in any case of polygenic or polyallelic contribution to a phenotype (if it's not a single gene causing the phenotype, it is incorrectly labeled as non-Mendelian).In most cases, non-Mendelian inheritance seems to be used to refer to inheritance of phenotypes, rather than inheritance at the genetic level. In other words, imprinting is considered a non-Mendelian phenomenon, because expression of inherited genes is based on parent of origin - even though at the genetic (sequence) level an intact allele is inherited for each autosomal gene from each parent.I question the idea that horizontal gene transfer has any implication for whether previous or subsequent inheritance is Mendelian or non-Mendelian. By example, if I engineer a knock-out allele in a mouse's genome, the null allele will be inherited in a Mendelian manner - even though the allele was from an exogenous source...My point: Horizontal gene transfer would seem to alter the genome content, but not the inheritance of that content.Perhaps I'm delving too much into semantics, but I'm not sure what you are after here - for example, work on transposable elements seems to be a hot area right now; would you consider those a mode of non-Mendelian inheritance?Another issue - do asexually reproducing organisms really follow Mendelian inheritance?And to clarify, do you think that HGT (or other specific findings) violates the modern synthesis somehow? HGT just seems to be one of many forms of mutation/alteration of a genome...[As an aside - I agree with you on the nature of ID popular books vs. evo primary literature; it seems many accuse the evo camp of somehow being weak because it doesn't simply quote popular evo books by mainstream 'experts', while any scientist knows (except perhaps Salty) that quotes don't mean a damn thing without objective evidence, except perhaps as defining historical context...]

From my readings the term "mutation" has expanded greatly to include HGT. Loosely, anything that changes the inherited genome is considered to be a mutation. In the case of the bacteria in your first post, it can be considered random since the bacteria is taking in ALL DNA from the surroundings. If this were a case of directed mutation then the bacteria would only let in DNA that would be beneficial and exclude DNA that would either be neutral or detrimental.Eukaryotes do not take on DNA in the same manner. The closest mechanism that eukaryotes have, at least in nature, is retroviral insertion. This differs greatly from HGT, of course, but it may still play a role in evolution.I think the synthesis is how HGT takes place. It is a random event, but plainly a mechanism to acquire new genes. There are also mechanisms, such as mu elements, that increase mutations during times of stress. However, the mutations caused by this mechanisms are random with respect to benefice just as HGT is. These two mechanisms seem to have one thing in common, increase variation when encountered with negative selection.As to Mendelian genetics, this deals with inheritence. The inheritence of new genes still follows Mendelian patterns.

Your last question: As far as I know 'NO'. How does one begin to construct a model that equally represents every single homo sapien permutation, or do we go about the approval process like the FDA 'beneficial to at least 5-7% of the test population? doesn't kill you? ta-da you're approved'.My following points are made in the context of complex diseases.
1) Evolution scientists like to use 'Neutral DNA Model'...this is crazy for the rest of us, BUT right now there is not a workable alternative...neutral DNA?? So, in our group we plug away at epigenetics and functional genomics (NOT the computer versions), and wait.
2) Mendelian genetics is fine if you are trying to explain "easy" stuff, but with humans if you try to publish a paper in a reputable journal and you say 'major allele' you better qualify that with ethnicity/age/geographical location (possible reason why a consensus for DNA model for evolution can't be reached and is certainly not trivial).

Hmmm. Would you consider Margulis' newest radical symbiogenesis hypotheses "non-Mendelian"? They're certainly non-Darwinian. I haven't run across any technical articles on her theories, but she and Sagan have published quite a few "popular science" books and articles espousing it.edited to add: someday I really must get a copy of "Acquiring Genomes" and read for myself how "radical" the ideas are.This message has been edited by Quetzal, 10-19-2004 09:17 PM

I'm not sure what you mean by this -How does Mendelian genetics fail in regards to complex traits?Mendel used simple binary traits in his experimental system - but Mendelian inheritance occurs with highly polygenic traits as well as monogenic traits.

Not previous or subsequent inheritance perhaps but the initial introduction into the genome is itself an inheritance, albeit from an exogenous source, and therefore distinct from de novo mutations of the genome based on changes or rearrangements of that genome in isolation as it were. No I wouldn't consider transposable elements moving around within a genome to be a form of non mendelian inheritance. I would consider retroviral insertions of exogenous genetic material to be.I have to point out that I am not saying that any of these processes must be both non-mendelian and non-darwinian, just that examples which can be charactised as either of these have arisen and may not as yet have been properly integrated into modern evolutionary thought.I was thinking along the lines of epigenetic effects such as DNA methylation, which is not only responsible for your imprinting but for other heritable changes, or heritable histone alterations of chromosome structure. Many of these epigenetic factors may not be heritable in the long term, at the moment it is not clear to what extent epigenetic factors are generally heritable or how stable those factors are. Also the existence of genetic elements which affect segregation and work to skew population ratios of certain traits, i.e. the Sex-ratio drive gene in Drosophila simulans. I don't think it violates the modern synthesis, but I feel that if you broaden things to the point where 'mutation' covers every concievable route for the introduction of changes in the genome, either genetic or affecting expression epigenetically, then you are making the term meaningless for the purposes of molecular genetics. Distinctions need to be made as to the operation and subsequent effects of these various mechanisms. Certainly they can all act to produce variation within a population and if you are only looking at the level of the phenotypic traits an organism presents then it may not matter what the source of that variation is, but I think that it is still important to distinguish between these various mechanisms as they pertain to the reshaping of a genome if we want to be able to reconstruct an organism or species genetic history as accurately as we can.TTFN,WK

I would consider endosymbiotic acquisition of genetic material, compartmentalised or not, to be non-mendelian, yes.TTFN,WK

quote:

---

No I wouldn't consider transposable elements moving around within a genome to be a form of non mendelian inheritance. I would consider retroviral insertions of exogenous genetic material to be.

---

I am not sure I get why they are qualitatively different. The end result is a proviral-like integration that was previously not present regardless of its origin. The only difference is that the origin of a within genome transposition would coalesce to a specific LINE, SINE, HERV etc. within the genome and the exogenous virus would be more similar to one of the retroviral classes. However, ultimately, HERVs for example are more similar to exogenous retroviruses than to other portions of the genome.In terms of revisions of the great synthesis, it is my impression that the Evo-Devo field is hoping to make such a breakthrough. That field of course, would have to take into consideration everything from mutations in DNA, imprinting, the histone code, etc.

Unfortunately, this is not the case. If it was then we would not need 50 million (at least, hehe) different medications for the same phenotype. Why? Say you have an individual who is homozygous at a certain position (just to make it easy). This genotype is then found to be in LD (linkage disequalibrium sp?) with 3 other positions in the same gene. However, each of these four have different associations with phenotype (there is some overlap). These phenotypic associations involve different genes. With just that do you think you could make a sweeping statement about the world popuation? I would hazard the answer is 'no' because one now needs to factor in ethnicity, yes it does matter, geographic location, and environment that these genes matured in. However, you should be able to if you were to think Mendelian Genetics can cover it all. At our center we have looked into how many different combinations one would have to consider if using mendelian genetics....the number gets very high. So, that's what I meant about "easy" versus what one gets in reality when studying complex diseases in humans. I certainly don't think Mendelian Genetics is defunct....quite the opposite...it's just not enough for us (our work). Therefore, in terms of what WK mentioned I think that the idea/scope of Mendelian Genetics needs to be overhauled. The whole idea of it referring to eye color/hair color/leaf size needs to be left in 2nd grade and more mature, realistic definitions of "traits" and concepts in high school and especially college need to be presented. Of course, weed out opinion from fact at your discretion.

It occurred to me that we really need a working definition of what might be considered 'non-mendelian' and 'non-darwinian'. I'll try a couple of definitions first and see what the rest of you think.Or that was what I planned to do, but between my difficulty in framing a really useful definition of either term, having started writing this post at around 10 this morning and only managed about half a paragraph on non-mendelian genetics by 5 in the afternoon, and actually having to do some work today I have totally failed to come up with any as yet, rest assured I shall give it some more thought and get back to you. If anyone else has any definitions they are partial to then all donations are gratefully accepted.TTFN,WKP.S. Here is what I did so far, any critiques or comments are welcome, unless you disagree with me in which case I shall have a hissy fit. This message has been edited by Wounded King, 10-20-2004 12:09 PM

This is indeed a difficult defintion to construct as it seems to depend ASSUMING no difference of infinte and finite ratios for any sign of the word signifers IN THE DEF., how NORMAL FORMS would normalize the data that traditionally is discussed under the word(s), "Mendel Ratio" (3:1) etc. and this seems to me differently ACCOUNTED for by Gould and Dawkins comparatively if sufficiently definable/defined. If one DID NOT THINK OF INFORMATION in this sense but only the relation to ENTROPY, I think accomplishing the writing of such defintions would be propositionally easier. I have not begun to figure out HOW EXACTLY predication IN KANT would be changed by such objectivication so thought. GOOD POINT WK! I myself get stuck because of the evolutionary thought experiment Ameisen gave as to cell death and mitochondria.for reference see
IIS 8.5 Detailed Error - 404.0 - Not Found
content etc
http://www.nature.com/cgi-taf/...
{Shortened display form of URL, to restore page width to normal - Adminnemooseus}This message has been edited by Adminnemooseus, 11-01-2004 10:47 PM

Taqless,Complex, or multigenic, traits are still inherited in a Mendelian fashion. It seems like you would consider a disease like cystic fibrosis as a 'Mendelian' disease, and one like schizophrenia a 'non-Mendelian' disease.However, as long as all of the individual alleles contributing to schizophrenia segregate according to Mendel's laws, it remains a trait resulting from Mendelian inheritance; as WK states above: You write: This seems to be your own personal hang-up though - Mendelian inheritance isn't limited to simple traits that can be followed with a Punnett square.Also, when I was taught Mendel's laws my textbook included human height to demonstrate highly polygenic traits. First of all, ethnicity doesn't matter, genes and environment do. Ethnicity happens to be defined by genes and environment, so it is often used as a control, however sloppy a categorization it might be...Believe me, I've seen my share of human disease susceptibility QTL analysis, but what exactly does this have to do with Mendelian inheritance? Mendel's laws do not state (that I am aware of) that the genetic influence on a trait must be constant in all subpopulations and environments.So I'll ask you, when dealing with complex human disease genetics, what mechanistically is different about the mode of inheritance of the genes?This message has been edited by pink sasquatch, 10-20-2004 02:30 PM