4 replies to this topic

[manofsan]

Please read this:

http://www.newscient...p?id=ns99996561

So how can so few genes make such complicated things as ourselves? On the bright side, perhaps that will make the genetic puzzle easier for us to unravel.

If only we could simulate the genome in a big computer matrix, to model the development of the human body in silico. Then we could play around with all of the genes or combinations thereof, to find out how to make different kinds of people.

[Cyto]

X-Ray crystallography can only give a picture during a specific moment in a protein's true nature. There are proteins that can fold 1, 2, 3 and more ways - also there are partial refoldings. (Check out: http://www.molmovdb.org/molmovdb/) I have to also throw in that a gene can skip an exon via influencing elements which can alter the protein. Multiple ORF starts for proteins can also contribute to changes. And with all of these forming tons of quaternary structures we should see quite the spread.

But for a computer to simulate the oscillations of all proteins is a tall order...

[kevin]

The ability to create a more complicated organism such as a human with so few genes while rice, a considerably simpler organism has so many more, comes from the regulation of the expression of the genes and in how they cooperate. If you take into consideration that 'non-coding' regions, or Junk DNA, (the latest study posted not withstanding), play an important role in development and regulation as well we can increase the complexity even more. Additionally, there may be only 20,000 genes, but how many splicing variants can be created from them and their regulation? The permutations are more than adequate.

[manofsan]

kevin,

I think we need to identify exactly what junk DNA does, by creating junk-knockout mice, and other higher level organisms. Engineering organisms without junk, or with modifed junk, will be the only way to truly know.

What about junk splicing? Also, can techniques used for gene investigation be used to investigate junk? If intron sequences can be grouped into coding sections called genes, then can exon sequences be grouped into any sections of relevance? Or can we only look at junk wholistically?

If junk is relevant, then it needs to be understood. Which means it needs to be tinkered around with, to see what makes it tick. Do epigenetic regulators bond to junk? Is junk merely useful as an overall contributor to chromosomal geometry and relational placement of genes?

Is DNA folding or geometry significant to its expression and regulation, the way that protein folding and resultant protein geometry is for proteins?

Answers, please?

[kevin]

A recent article (http://www.biomedcen...ews/20041011/01) describes how transposons can regulate development and it is known that they can exist in junk DNA.

I wonder if one potential use of 'junk DNA' is a storage place for transposon sequences which are used as elements of regulation when they get placed into promoters?

I'm not sure if DNA folding is important to gene expression or regulation as there doesn't seem to be sequence specific variation in nucleosome binding or chromatin structure (as far as I know yet.. ) What I have been reading is that the availability of binding sites for transcription factors (and presumbably other genetic agents like RNA based enzymes), as determined by the position of the sequence as wrapped around the nucleosome core, can have some bearing on the level of transcription. Binding sequences located closer the beginning of the wrapping tend to be more available as they are not as closely associated or tightly bound and are thus more accessible.

I'm sure higher order structure plays a significant role by way of nucleosome-nucleosome interaction, possibly through the modification of epigenetic information but whether or not there are subtle variations in the actual folding of DNA which contribute to gene expression I think is undetermined at this point.