DNA Coding

Discussion of anything and everything that happens within the Iris Alternate Reality Game.

Moderator: Moderators

Ceantari
Data [Conditional]
Posts: 30
Joined: Wed Aug 08, 2007 12:27 am

Re: DNA Coding

Unread post by Ceantari »

beelzebub wrote:
mtwain wrote:The genetic code we were given was for the Chinese Alligator's mitochondria, correct? If I'm not mistaken, mitochondrial DNA doesn't change very quickly... has anyone tried checking this code against human mitochondrial DNA? The message was definately intended for a human audience, whatever that message is -- it makes sense that if it's to be compared to any genetic code, it would be human.
This is wrong. Mutation rate in the mitochondrion is much higher than in the rest of the genome, in mammals for sure and probably in most other vertebrates. At any rate I compared this sample sequence against the American Alligator's mitochondrial genome (available here), and it only has 83% similarity. It's definitely built from the Chinese Alligator sequence. Nothing else in the database is even close.
Also, doesn't really help that there are only 2 alligators in the world: the American Alligator and the Chinese Alligator. :/
beelzebub
Data [Conditional]
Posts: 48
Joined: Thu Jun 14, 2007 6:08 pm

Re: DNA Coding

Unread post by beelzebub »

sassafras wrote:
beezlebub wrote:Not to start a flame war here, but mtDNA is much more highly conserved than chromosomal DNA. We use mtDNA to tell the relationships of vastly different species when a traditional chromosomal analysis would be completely irrelevant. You have to understand that the bulk of the Eukaryotic genome is actually non-coding or relic genes that are nonfunctional (called pseudogenes). Because of the high improbability of DNA in this region actually being significant, many more changes occur here that are not corrected through time because they have no selective advantage, and therefore Darwinian Natural Selection does not come into play and these mutations accrue rapidly.

mtDNA on the other hand is more bacterial in makeup, meaning that almost the whole loop strand is a coding gene of some sore. Because it is so highly functional, mutations in mtDNA are ~99.5% of the time, deleterious to the organism and therefore they get "weeded out" quickly. Consequently, mtDNA is much more conserved throughout time than chromosomal DNA.

What am I getting at? Well, an 83% correlation between human mtDNA and alligator mtDNA is not out of bounds, it is about right, considering that branch of evolution is probably close to 120 million years ago (when the squamosa diverged into what is currently mamallia and reptilia). So my point? This isn't human in any way, but it could very well also not be crocodile. A 20-40 bp difference is significant in mtDNA. That's not random variation, that's sister species or sister taxa at least.

...Besides, I think this is just like the star chart. They took it from an actual source, changed it a little so it would be unique to the game, and then we find the original and obsess about that significance until we get another clue.

To continue the flame war you started, this is not what I said. i said the mutation rate in mitochondrion is higher than in the rest of the genome. This has no bearing on whether more is conserved because of selection; it is simply true. The mitochondrial DNA polymerase is crappy and has lower fidelity than the nuclear polymerase. Second, the reason the mitochondrial genome is used in taxonomic studies is because (a) it is small and easy to sequence (b) it is inherited from the mother and so has an unambiguous descent (c) it is easily aligned, since it only has a handful of genes and (d) it is ubiquitous in eukaryotes. The mammalian genome, by comparison, is much, much larger and has many duplications. It's these complications that make it less suitable. Otherwise you'd be perfectly fine doing taxonomic classification using, say, the alcohol dehydrogenase gene.

But, getting back to conservation, you're also wrong there. It's simply not true that the mitochondrial genome is better conserved than even the "junk" parts of the human genome. You can satisfy yourself of this by BLASTing the chimp mitochondrial genome against the nucleotide database. The accession number is NC_001643. You'll find that identity with human is only 91% - far lower than genome-wide identity. The reason is that mitochondria do not undergo recombination, being inherited from the mother only, so selection is much less efficient at removing mutations. They're also subject to something called Muller's Ratchet, which you can read up on.

Finally, the 83% figure I quoted was for identity between the mt genomes of A. mississippiensis and A. sinensis, not with human mt. Meaning, the sample.txt DNA ONLY has 99% identity with A. sinensis, and only that could be the source.

But, all that said, I agree with you - I don't think there's anything more to see here. It would be a good puzzle, but not the sort of thing we've seen so far in Iris.
sassafras
Data [Conditional]
Posts: 36
Joined: Tue Jul 10, 2007 1:37 am

Re: DNA Coding

Unread post by sassafras »

beezlebub wrote:
To continue the flame war you started, this is not what I said. i said the mutation rate in mitochondrion is higher than in the rest of the genome. This has no bearing on whether more is conserved because of selection; it is simply true. The mitochondrial DNA polymerase is crappy and has lower fidelity than the nuclear polymerase. Second, the reason the mitochondrial genome is used in taxonomic studies is because (a) it is small and easy to sequence (b) it is inherited from the mother and so has an unambiguous descent (c) it is easily aligned, since it only has a handful of genes and (d) it is ubiquitous in eukaryotes. The mammalian genome, by comparison, is much, much larger and has many duplications. It's these complications that make it less suitable. Otherwise you'd be perfectly fine doing taxonomic classification using, say, the alcohol dehydrogenase gene.

But, getting back to conservation, you're also wrong there. It's simply not true that the mitochondrial genome is better conserved than even the "junk" parts of the human genome. You can satisfy yourself of this by BLASTing the chimp mitochondrial genome against the nucleotide database. The accession number is NC_001643. You'll find that identity with human is only 91% - far lower than genome-wide identity. The reason is that mitochondria do not undergo recombination, being inherited from the mother only, so selection is much less efficient at removing mutations. They're also subject to something called Muller's Ratchet, which you can read up on.

Finally, the 83% figure I quoted was for identity between the mt genomes of A. mississippiensis and A. sinensis, not with human mt. Meaning, the sample.txt DNA ONLY has 99% identity with A. sinensis, and only that could be the source.

But, all that said, I agree with you - I don't think there's anything more to see here. It would be a good puzzle, but not the sort of thing we've seen so far in Iris.
We're using the same evidence to argue opposing points. Yes, I know about Muller, and yes I know about all of the points you brought up. As it were, I argued my point using exactly contradictory evidence to what I was saying. To think, I have a degree in Molecular Biology and I rearranged the fundamental idea of inheritance. Meh, I always hated mtDNA anyway.

You are right, it's great for studying over a few hundred generations, but breaks down after that because it tends to accumulate quite a bit of variation. It's the duplications in the eukaryotic chromosome that make for great long term study. And actually, ADH is generally a great long term tool because it's an essential protein that is very conserved and has an obvious promoter region that we can build a good primer for when doing the polymerase prior to sequencing. Precisely the opposite of what I was saying. I read my old post and felt sort of stupid, because, had I thought about it I wouldn't have posted it.

Back to the point though, I still think the fact that it is a slightly rearranged sequence of alligator to be irrelevant. I don't know that we will glean anything more from this evidence than we did from the server one star map. In reality, only one piece of evidence from each server has actually led to unlocking the next (with the possible exception of the current, because the boomerang nebula seems to be a second, necessary, hint). I just think we're supposed to start thinking about DNA sequencing, not solve the code.
thereIwasn't
Data [Authenticated]
Posts: 203
Joined: Wed Aug 08, 2007 10:25 pm
Location: Somewhere (not there though).

Re: DNA Coding

Unread post by thereIwasn't »

Just to ask, has anyone found anything more amazing or groundbreaking than the fact that what we're looking at is a slightly messed up chinese alligator? I must admit, I'm tired of hearing all of these 'flood infested alligator' theories.
zoned out
Data [Conditional]
Posts: 33
Joined: Fri Aug 10, 2007 7:39 pm

Re: DNA Coding

Unread post by zoned out »

thereIwasn't wrote:Just to ask, has anyone found anything more amazing or groundbreaking than the fact that what we're looking at is a slightly messed up chinese alligator? I must admit, I'm tired of hearing all of these 'flood infested alligator' theories.
Agreed.
Post Reply