Protein folding, the origin of novelty and everything

Protein folding finally gets its own thread! Let no-one ever again think of genes as simply lists of A, G, C and Ts!Here is a great mainstream web site explaining protein folding and protein folds (we have talked a lot about this in various threads around here): http://www.lmb.uni-muenchen.de/users/steipe/lectures/structure/node03.htmlFor example we see that estimates put only about 1 in 10,000 random sequences folding at all and of course these would be non-functional. So a new gene starting from random DNA (or drastically altering fold from an existing gene) has to fight this sort of barrier even before it can begin to be selected for. After chancing on to a foldable sequence it still has no function so it will drift away from the foldable sequence.PS - for any protein newbies here, proteins are what genes code for and they make the world go around by doing all of the structural, catalytic, signalling, detecting, transport, motor and metabolic funcitons of the cell either directily or after making a non-protein chemical to do the job. They only do this job if they have (i) a foldable seqeunce, (ii) have an active or catalytic site and (iii) this biochemical function is useful in some cellular context.On this web page you can find some beautiful schematics of protein folds (scroll to bottom): http://www.lmb.uni-muenchen.de/users/steipe/lectures/structure/node04.htmlHere is the web site where structural biologists deposit these 3D datasets determined by X-ray crystallography and NMR (predciton from seqeunces is an ongoing struggle but progress is being made): www.pdb.orgalthough here is probably a better site where one can peruse all of the known folds via the "CATH" catalog: http://www.biochem.ucl.ac.uk/bsm/cath_new/class.htmlHere is a rather impressive prtoein fold (click on image for a larger view, alpha-helices are magenta and beta-strands are yellow): http://www.biochem.ucl.ac.uk/bsm/cath_new/domains/3aahA0.html(The amazing thing is that just about every individual molecule with that sequence will fold to that exact shape so that it can do a very specific job.)Here's another fold: http://www.biochem.ucl.ac.uk/bsm/cath_new/domains/1lxa01.htmland one more: http://www.biochem.ucl.ac.uk/bsm/cath_new/domains/1rie00.htmlAnd here's how the cell makes proteins from RNA copied from DNA on a gene: http://www.accessexcellence.org/AB/GG/protein_synthesis.htmlThe ribosome that reads the RNA and strings the amino-acids of the protein together (green in the previous link) is itself an association of about 50 proteins and RNA: http://www.molgen.mpg.de/~ag_ribo/ag_franceschi/So sometimes, the ribosome unknown to itself, is making more of itself. Here is a rotating model of the 50S half: http://www.molgen.mpg.de/~ag_ribo/ag_franceschi/franceschi-projects-50S-2.htmlEach of the colored bits is a folded ribosmal protein required to make the ribosome translate RNA into protein. The gray is structural RNA (not the RNA that is translated). Here is the other half of the ribosome complex: http://www.molgen.mpg.de/~ag_ribo/ag_franceschi/franceschi-projects-30S-2.html

​

[This message has been edited by Tranquility Base, 07-16-2002]

^ I agree selection is not random but the generation of sequences to be selected is so the 1 in 10,000 holds. The point is that before selection can even start to come into play the estimate is that there is a 1 in 10,000 hurdle. This statement is not reversed by the fact that 10^10 seqeunces can fold to the same topology. 10^10 is nothing compared to the 20^100 possible sequences!The point is that there is a very sharp discontinutiy between mutations optimizing existing genes and generating new folds! it is the reason why we don't see new protein families in the lab or in the field. It is the reason why no-one expects the Galapogos finches to be dsintinguishable by protein families. It doesn't make macroevoltuion impossible automatically but it shows that evidence for microevolution is almost irrelevant.

​

[This message has been edited by Tranquility Base, 07-16-2002]

BradThe all folds estimate comes from the ones we've seen and looking at how much of the human genome is left (with unknown folds) so it is not from biological necessity but empericism. It is like sampling a poulation and estimating the result if we could exhaust the population.As usual I don't understand about half of your post!
1. What is the ICR modelyou are talking about?
2. How could eletrons and metals have anything to do with protein folding?
3. What is reproductive doubling?