Manipulation of DNA by cells?

Depending on what exactly structure of DNA means in this sentence, yes, definitely - as an absolute fundamental part of Eukaryotic biology.Simplifying a little: DNA is a long thin molecule, in normal cells it is wrapped and coiled up with proteins into chromatin. How loosely or tightly this chromatin is packaged is a fundamental part of gene regulation. In fact, it's coming to be seen as the major unsolved part of how gene regulation works, probably as important as transcription factor regulation.What's more a significant part of the control of how chromatin is arranged is through DNA modification via methylation (a semi-heritable form of epigenetic control).So, yeah, cells modify the structure of their DNA both chemically and physically.In any case, this:

quote:

---

Thus the cell is no longer a Cartesian duality, meaning separate molecules as information carriers, and a separate set up for carrying out that information. Or the whole cell participates in the cellular process.

---

is wrong anyway. Early work on DNA may have thought that, but it's well understood that the idea of passive DNA is wrong although unfortunately widely spread through poor analogies such as "DNA is the blueprint for life". DNA has a key role to play in the ongoing control of biological activity, it is not a passive molecule that simply make things that then go do their stuff. editOh, and - of course - transposons. There are parasitic* elements called transposons that exist within the genome that code for proteins that will cut them out of the gene and re-insert them elsewhere or make copies of themselves. These can occasionally also move other bits of DNA around.* - some thing that transposons have an adaptive purpose, I think they're merely parasites. I think my view is probably more common.Edited by Mr Jack, : Added transposons

What do you mean by a purposeful process?It isn't just something that happens, it's a vital part of how the immune system functions. The random recombination produces a "library" of antibodies that will bind to almost any pathogen.

A random manner; that's the brilliance of the system. If it worked in a fixed manner it would only work against pathogens previously encountered at sufficient levels to drive evolution of immune responses to them. By randomly recombining elements it covers a much greater epitope space.

Say I want to generate a random registration number. It consists of two letter, the numbers 0, 1, 5 or 6, another number 0-9, and a sequence of three letters.I get my bag of scrabble tiles out for the letters, and pull out two. Then I take some Rummikub tiles pick out just the numbers 1, 5, 6 and 10 (counting 10 at 0), put them in a bag and pick one out a random. Next, I pop the numbers 1-10 in and pick out another, finally I pick three more scrabble tiles out.Wow: BH07 LUA - I've just generated a numberplate at random, but all the letters and numbers are in the right order and its a perfectly correct and valid numberplate.Things can be random while still being structured. Random does not mean anything goes.

If I make enough of them, yes, it will. And this is, essentially, the principal behind the immune system. Random recombination produces vast numbers of different antibodies, so that when a novel pathogen enters the body chances are that one of the vast number of random antibodies will recognise the pathogen. The body then identifies this fact and mass produces the antibody.

The body's immune system doesn't carry any specific defences against any pathogen by default. It simply produces a sufficient range of antibodies to detect almost anything. Antibodies can be raised to recognise almost any protein (actually a small part of it) based on random recombination of a small number of pieces.I'm afraid I can't remember off-hand how the body identifies and clones the antibodies needed in response to an attack. Hopefully another poster can fill you in.