Gene methylation

[superluminal]

Genes basically code for protiens in the way they are expressed. Certain genes are "methylated" which modifies their ability to be expressed. Heavy methylation apparently stops expression of the gene, while less methylation just decreases it's rate of expression. This then is a form of "metagenetic" information. Heritable information that is not a gene.

- What does this say for the human genome project results?

- Do we only have a part of the story?

- Do we need a "methylation project" to really understand how it all works?

If this has been discussed, sorry for the repeat.

 

[CharonZ]

- What does this say for the human genome project results?

Not much. The Hugo project aimed to obtain the complete DNA sequence, which will offer a sound basis for further investigations. Epigenetic functions can of course, not be derived solely from this info, but that was never the aim in the first place.

- Do we only have a part of the story?
Of course. The raw sequence was never meant to answer all questions.

- Do we need a "methylation project" to really understand how it all works?
We can even go further. Especially regarding regulation we also need to research protein-DNA,Protein-Protein, RNA-DNA-interaction, methylation etc.
Furthermore the exon-intron structure of eukaryotes makes the predictions of genes itself rather complicated. And even now, several years afte the complete sequencing of the human genome, it is not completely annotated as yet.
To summarize it, we are still at the beginning of exploiting the information that is available through the HUGO-Project and further studies in all directions are still needed.

 

[superluminal]

CharonZ,

Thanks. It's all much more complex than its presentation to the general public would suggest, isn't it?

If you can, would you (or someone) briefly describe how all of this meta or epigenetic structure gets replicated? I understand that the helix itself is reproduced by "unzipping" down the middle, and a theoretically identical copy is made by attaching complmentary bases to the exposed ones, drawn from raw materials inside the cell (by catalytic enzymatic reactions?).

How are a particular genes methylation sites, and degree of methylation determined?

 

[zyncod]

You have hit on one of the fundamental questions of biology - that no-one has any real answers to. Most methylation occurs in CpG islands - a cytosine followed by a guanine. This methylation uses enzymes - DNA methyltransferases that transfer a methyl group from an amino acid to the cytosine. And this process is heritable - genes that you get from your mother that were methylated are methylated in you as well. But no-one knows how the methyltransferases "know" that the strand that they're replicating is methylated as well. So the replication of methylation patterns is kind of a mystery.

A lot of people think that single copies of genes are methylated in order to fine tune the level of protein coming from that gene (less mRNA = less protein). Also, methylation can 'silence' useless DNA, inserted viral genomes, and transposons. That process is actually kind of cool - the methylated DNA brings in other proteins that 'twist' the DNA tighter so that proteins (such as polymerases) have a tougher time accessing it. But, again, no-one really knows what the signal is that says: this piece of DNA (say, a viral genome) should be methylated. Although people doing work with RNA interference (RNAi) are finding out that these two processes may be related.

 

[zyncod]

And, yes, in the post-genomic era, there a lot of other "-omes" that we have very little information on - the proteome, the metabolome, the transcriptome, the glycosylome, and on and on and on. And on.
http://www.mindfully.org/GE/GE4/Nomenclature-Crisis26apr02.htm
But the human genome project's importance really can't be understated as all these other "-omes" would be almost useless without it.

 

[gendanken]

GRAHAHA!!

Also, methylation can 'silence' useless DNA, inserted viral genomes, and transposons. That process is actually kind of cool - the methylated DNA brings in other proteins that 'twist' the DNA tighter so that proteins (such as polymerases) have a tougher time accessing it.

You see???!!
Sing it, brothersister. 'tis no asstalk.

 

[Hercules Rockefeller]

But, again, no-one really knows what the signal is that says: this piece of DNA (say, a viral genome) should be methylated. Although people doing work with RNA interference (RNAi) are finding out that these two processes may be related.

In what way are you proposing that RNAi and methylation are related?<P>

 

[zyncod]

The siRNAs produced by Dicer or other components of the RNAi pathway might bind to methylation proteins and guide them to homologous regions in the genome. Here's a paper where they found out that knocking out RNAi genes abolished centromeric methylation.
I mean, it kind of makes sense when you think about the fact that RNAi silences dsRNA viral genomes. It would be useful to have a way of silencing proviral genomes inserted in the host genome - and the siRNAs would be a guide to what the proviral genome sequence is.
Reinhart BJ and Bartel DP. (2002). Small RNAs correspond to centromeric heterochromatin repeats. Science 297:1831.