|
Author
|
Topic: "Junk DNA" as Cannon Fodder
|
Art
Member
Member # 179
|
posted 09. September 2002 23:24
Interesting comments, Leonard, that bring to my mind a few questions:
1. What do the cells of Lilium longiflorum possess that is absent from those of Arabidopsis thaliana, such that the former has a need for almost 10^11 bp of "junk DNA"?
2. Many models of the eukaryotic genome actually hold that it functions as a circular entity - with the "ends" being held together by the various nucleus-localized complexes.
3. It is quite likely that energy-providing intracellular membranes in eukaryotes pre-date eukaryotes. Some sort of fusion of your ideas with those of Margulis might be an interesting direction to explore.
IP: Logged
|
|
Leonid Andreev
Member
Member # 282
|
posted 10. September 2002 13:11
Art:
1. It would be irresponsible of me to even attempt answering a question like that. Not only does any plant or animal species have its own history, but it also trails its evolution history which is tightly connected with the evolution of all the living matter on Earth. A species that emerged in an exceptionally favorable environment might have further gone through a harsh ecological pressing, which might have made its survival depend on its metabolic plasticity. That is exactly why genomes of many species of higher eukaryotes include an enormous amount of genetic fragments of viruses and bacteria. Investigation of detective files is not exactly my hobby. Any ideally designed species might have ended up being a very complex palliative of an original design, which by no means makes it a failure once it survived to the present times.
2. I must apologize for my inaccurate wording. I should have said that a eukaryotic genome has difficulties in functioning at a circular DNA level. The difficulties correlate with a size and location of repetitive DNA.
3. Margulis’ conclusions on bacterial origin of mitochondria are the result of her interpretation of experimental data obtained by herself and others. My hypothesis of the origin and evolution of life is built mostly on a theoretical basis, and, therefore, the agreement between my hypothesis and the views on the origin of mitochondria is even more so ponderable.
IP: Logged
|
|
Janitor@MIT
Member
Member # 125
|
posted 13. September 2002 12:25
I don’t want to distract from the discussion of John Bracht’s interesting musing, which appears to me to be sound if target profile dimensions are the only relevant factor. As an (lengthy) aside, we should beware conflating the “target theory,” intended only, or at least originally, as a method for elucidating the spatial dimensions of the gene, before it was chemically isolated, with a theory of evolutionary mutagenesis. Many experts argue that radiation induced mutagenesis carries far less weight than mutagenic factors in “housekeeping” operations, duplication, and the intracellular environment. This is supported by a simple and intuitive argument: life forms have evolved for ~ 4 billion years in an environment with background radiation. This background radiation then becomes a very elementary and ineliminable factor in their design, in a sense, just as gravity is. In other words, we should expect them to be well-adapted to such factors, and factors intrinsic to the design, such as chemical reactions (cellular operations) occurring in a solvent at ambient temperature, are likely to be more prevalent mutagenic factors than any factor “extrinsic” to the design (such as solar radiation). Obviously, this is a controversial conclusion. And to anticipate, the argument is not that radiation does not cause mutation, but that radiation as a factor in the design-evolution may be somewhat exaggerated. (A Cold War hangover?)
Very simple principle: information is “lost” or altered in performing operations upon it. Obviously, keeping it “shielded” when operations are not being performed upon it helps to reduce the load, but a “cost-benefit” analysis has to be made. Given that degradation is inevitable, does the benefit of carrying significantly more DNA outweigh the cost? We should keep in mind when considering cost-benefit, that organisms do have quite efficient error-proofing systems in place and one of them, complementarity, already doubles the amount of DNA. (Sometime in the mid-1960’s Charles Bennett actually did some cost-benefit calculations along similar lines but wrt specifically to the requirement of unique decodability. If I recall correctly.)
Something else to consider, John Bracht, your musing appears to be based on the traditional conception of a gene as a sequence of DNA encoding the structure-function of a protein and those elements directly controlling the expression thereof. (That last part, control, needs to be considered in its full implications, which I’ll do in part below.) Maybe the first step to be taken in accounting for much of the rest of DNA is as simple as expanding our definition of what a “gene” is. Obviously, the distinction between coding and non-coding (“junk”) sequences rests upon the presumption that there is only one DNA code, or the already falsified idea that only protein structure-function is what is encoded.
Along these lines, some theorists have suggested the existence of more than one DNA code. I find this idea particularly intriguing (exciting actually), not the least reason for which, is that I have also made the same argument on design theoretic principles. Code developers design de novo and adapt existing codes to different functionalities. No reason why this shouldn’t be true in biological design-evolution as well. We know it happened once—the proteome code. But until recently, no evolutionary biologist considered that it may have happened twice (or more). Traditional theoretical presumptions impose a false implicit parsimony here: it’s difficult enough to account for the production rules of one code. Why make the problem even more complicated than it already is? Well, one reason to make it more complicated is the existence of “junk” DNA and the difficulty of accounting for it on any general theory that presumes only one code-functionality.
(What I’m suggesting is not really an “alternative” code as usually conceived, such as that often found in ciliates or mito-codes. There is a simple test of the existence of “alternative” codes, which I’ve suggested elsewhere, but I received nothing but negative feedback on it: For a code to be truly “alternative” requires an expansion or contraction of the “canonical” translation table—adding or removing rows or columns from the table, not merely “juggling” the entries. This is because the “canonical” code is a hybrid—it has a “statistical” property--adaptability on the basis of the DOF defined by statistically permitted “near-neighbor” transformations. But what I've suggested here is the possible existence of a code(s) that isn’t even based on the proteomic table, although, obviously, its elements, nucleotides, are the same. It may not “translate” for anything, but instead mediate operations, if you know what I mean.)
Simply on the principle of design economy I find it difficult to believe that almost all DNA is non-functional. The same design principle applies to genomes that do not have much “junk” DNA, so it might be productive to consider the different demands placed on genomes that do contain large tracts of non-proteomic DNA. Since significant amounts of “junk” DNA are found in euks (particularly metazoans) we might consider what they do that proks don’t. One of the obvious differences is that euks typically have more elaborate life-histories. DNA does not store information in three-dimensions only, but four. In a sense, euks are inhabitants of that fourth dimension in a way that proks are not.
Even though biologists find useful a computation-communication analogy, there is one consideration that is quite noticeable by its absence: timing, task scheduling, sequencing, queuing, idling, assignment, allocation, access, and communication and control delays. Given that typical cellular dimensions and time scales are tiny, this might not seem to be an important factor. Quite the opposite is true. Also a significant factor is that these processes are distributed and parallel, which represent unique engineering problems, esp. wrt program coordination, timing, scheduling, and delays. A “timing theory” is just an extension of the already proposed “spacer theory.” It might be helpful also to consider that the computation-communication analogy is not complete. Cells don’t operate only like nano-computers. They also operate like assembly lines. Etc.
John Bracht calls this “musing,” but all I’ve added to his musing is “wild speculation.” LOL Sure is fun, isn’t it?!
Just a parting shot: The remarkable resilience (“Invincibility” in the Dembskian sense?) of the dysteleological perspective (implicit in “junk”) in the face of the facts never ceases to amaze me. But then, it’s not really a theory as much as it is an “attitude.” Yet, there probably isn’t an attitude that has had such a poor track record in biological science. As near as I can tell, it’s invariably proven wrong as our knowledge increases. But some people insist upon it as a sort of default view! How about, instead of assuming that this “theory” of “junk” DNA, plainly proposed in ignorance, is true, we assume the opposite? Why not, instead of assuming that life forms are sloppily, even accidentally, assembled agglomerations, don’t we consider them exactly for what they are—exceedingly complex, well-integrated, smoothly and efficiently functioning systems of remarkable ingenuity and sophistication?
I suspect that this is really what all biologists assume and that the dysteleology is asserted for non-scientific reasons. It might just as well be dropped, as it never adds anything to our understanding.
IP: Logged
|
|
Leonid Andreev
Member
Member # 282
|
posted 15. September 2002 02:17
Janitor@MIT wrote:
quote:
I suspect that this is really what all biologists assume and that the dysteleology is asserted for non-scientific reasons. It might just as well be dropped, as it never adds anything to our understanding.
The way of thinking expressed in these words, is, in fact, is a germ of new "feedback" knowledge growing at the junction of two fields of science – pure biology and molecular biology. As a matter of fact, pure biology and pure molecular biology each avails itself of knowledge that is both complementary, in Bore’s sense of the term, and contradictory – as long as there is no appropriate sophisticated methodology to put together the knowledge available in both domains.
In early 80’s, many experienced and erudite biologists were knowingly letting their professional growth lapse for the sake of the new, prestigious path toward the “exact” knowledge. In the meantime, most of the papers, essays, and monographs on molecular biology used to sound like solo concerts of singers with perfectly professional voice and articulation – singing arias made of lines from a large collection of different operas (different areas of biology). (My critique, certainly, refers to the general tendency, not particulars).
Molecular biology is reductive in itself. Pure biology becomes obsolete as soon as an organism has been thoroughly (even hypothetically) investigated. Therefore, today, it is clear, as never before, that although optimism brought in by molecular biology is perfectly justified, the volume of work yet to be done in that field is about the same as percentage of “junk DNA” in eukaryotes. Only broad biological correlations based on adequate philosophy and methodology can reunite separated exons and introns as native DNA that functions in a living cell.
The above being an issue for separate discussion, I just want to mention an important advantage of biological correlations: they result from comparison based on impeccably selected (from a biological standpoint) sets of organisms, which allows one to advance step by step when making a conclusion on which of the outstandingmolecular-biological characters correlate with a certain biological function.
quote:
some theorists have suggested the existence of more than one DNA code. I find this idea particularly intriguing (exciting actually), not the least reason for which, is that I have also made the same argument on design theoretic principles. Code developers design de novo and adapt existing codes to different functionalities. No reason why this shouldn’t be true in biological design-evolution as well. We know it happened once—the proteome code. But until recently, no evolutionary biologist considered that it may have happened twice (or more). Traditional theoretical presumptions impose a false implicit parsimony here: it’s difficult enough to account for the production rules of one code. Why make the problem even more complicated than it already is? Well, one reason to make it more complicated is the existence of “junk” DNA and the difficulty of accounting for it on any general theory that presumes only one code-functionality.
An assumption of existence of at least two codes seems to be quite reasonable. Upon synthesis of mRNA, serving as protein synthesis template, the splicing technique is applied to a living cell for removal of “junk DNA” by cleavage and ligation reaction. This process is completed inside a nucleus, after which protein-coding sequences are expressed out of the nucleus. In the meantime, there have been numerous findings within the past decade that point to a facilitator role for introns. It is especially well expressed in fresh cells, whereas in old cell cultures introns role is not so critical. As J. Bergman notes, “Regulatory functions of introns may involve controlling gene activity in different developmental stages or responding to immediate biological needs by controlling local gene expressions. This function of introns could occur if, as one theory indicates, exons code for a domain, a polypeptide unit that has a discreet function such as binding to a membrane, or to the catalytic site of an enzyme or serving as a structural unit of a protein.” (www.rae.org/introns/html).
Responding to Janitor@MIT’s concern that “no evolutionary biologist considered that it may have happened twice (or more)”, I can only point out that this is exactly what my previous posting in this thread was about: a biologist’s view on how and why eukaryotes acquired the “new” code while the “old” code exists in both eukaryotes and prokaryotes (cf. page 2 of this thread).
[ 15 September 2002, 02:20: Message edited by: Leonid Andreev ]
IP: Logged
|
|
DChaves
Member
Member # 422
|
posted 22. September 2002 00:11
not an expert here, but some speculation on "junk DNA" or "vestigial sequences":
It seems that junk DNA may be important for many reasons:
-as per the earlier suggestion, it could provide "mutational hot spots" for protection of genes from mutation accumulation. (in calculating in vivo mutation rates it seems that gene mutation, moreso than junk DNA mutation, would predominately be associated with polymerase activity, rather than mutagens. is this taken into account? or what effect would this have?)
-providing additional regulatory elements, etc. (ex=recruitment of cohesin)
-enhancing DNA accessibility to transcriptional machinery based on nucleosome packaging. importance for HDACs and HATs in chromatin remodeling and transcription initiation.
-a role for chromosome stability and balance during meitoic divisions?
-a general effect on DNA supercoiling?
-any association between telomere length and levels of junk DNA?
-a role in designed evolution? i.e., a basis for inherent potentiality, or the creative ability of cells to respond to environment via genetic change. elements with low selection pressure that can accomodate substitutions and may be associated with designed micro (or macro?) gene evolution.
-or possibly, there is no functional role for "junk DNA". an artist may add an extra brush stroke not because it is necessary for an effect, but simply because it brings the painter pleasure.
can junk DNA hypotheses be tested with YACs? what about microarray analysis of mRNA production following any junk DNA deletion?
best regards,
dc
IP: Logged
|
|
|
Printer-friendly view of this topic