Teach me about DNA/RNA

Ben, how familiar are you with chemistry? What's the highest level chemistry course you've taken?

Without really wanting to break up the fun I have to mention that while not really wrong (at least just from speed reading it) there is some confusion in the above posts regarding certain concepts. It is especially worthwhile to note that certain things mentioned only apply to eukaryotic cells, but not to prokaryotic. Learning just based on random posts will result in flash-card knowledge. If you are really interested you really should take a peek into textbooks and then maybe ask specific questions.


It really might be more worthwhile to grab a textbook for starters rather than trying to make sense out of posts in a forum. The reason being that the topic is rather complex (if not complicated) and posts with a certain skewed focus might actually contribute more to confusion than clearing things up, in my opinion.

I'd like to say that biology is so full of exceptions, it's hard to find rules. But before listing every single exception, I thought we could at least give him a general idea of how stuff is working, some of the time. That way he could actually understand how the exception is an exception.
 
What does this mean?
With respect to Roman's previous comments, there is a general flow of information within each cell. The DNA in your nucleus has a copy of every single gene in our genome (two, actually, going back to the haploid/diploid discussion), but not all of them are being used at once. Once a cell specializes (e.g. bone marrow cells, brain cells, liver cells, etc), it will only need the genes that are specific to it's function. The DNA is transcribed into RNA, which is a complementary strand to the gene that is being transcribed. The transcribed copy of the gene is an mRNA molecule, which then travels outside the nucleus to the ribosomes, where it is translated to the protein end result. This is the general flow of information from what is stored in the DNA to a functional end product.

Reactive with respect to what?
1.) How does it "coil"? It seems that if it's, say, laying out on my table, it would need some impetus to "coil".
2.) Given that there are only a finite number of base pairs, it seems to me that large chunks of the strand (I'm assuming it's a strand) look more or less like other large chunks of the strand. Given that the base pairs bind with some strength (van der Waals forces?), and that the strand of RNA itself has some tension, I can imagine a situation where one large chunk of base pairs could be out of place, and the interaction between conjugate bases would be too strong for the tension of the RNA to overcome. Does this happen?
3.) Given the above, does it coil up the same way every time? What if it coils up wrong?
RNA is generally single stranded, but nucleic acids are more comfortable in a double-stranded conformation. So we notice that the strand will twist and contort so that it is able to form hydrogen bonds between bases elsewhere in the strand.

Each nucleoside, which is the nucleotide base attached to either ribose (in RNA) or deoxyribose (in DNA), is covalently bound at the 5th Carbon atom of the sugar molecule to a Phosphate ion. The phosphate is then covalently bound to the 3rd Carbon of the sugar in the next nucleoside (forming the sugar-phosphate backbone). The nucleotides are covalently bound to the 1st carbon of the sugar. This covalent binding of the nucleotide to the sugar backbone prevents mismatching in the way you are describing in part #2 above, because the strands of nucleic acids are fed through the transcription or translation machinery like ticker tape.

RNA will fold the same way each time under normal conditions, since the conformation they land in is the most thermodynamically stable. Under pH changes, temperature changes, etc, a more stable conformation may be found that looks different.

So the mRNA is just somehow a copy of the DNA, I guess? What is the mechanism for "copying", and does it happen at regular intervals? If so, how does the "copier" know when to do its job? Behind all of this is some chemical reaction, like some intricate swiss watch made out of carbon, nitrogen and oxygen---is this correct? What makes the whole process go?
mRNA is a transcription product, and is an intermediate in the flow of information. Transcription is handled through a multi-protein complex that binds to the DNA strand takes free nucleosides to create a complimentary strand. This complementary strand is the mRNA. The copier has many signals that it needs to do it's job, and a lot of them are not well understood. Ion concentrations in the cell's cytoplasm, hormone messengers that bind to protein receptors, the presence of a toxin -- chemical signalling, in other words.

Why would you want to modify it? And what does this even mean?
mRNA can be processed before it is translated into protein. This is useful because it allows a single gene in the DNA to be used for several similar proteins. Regions can be conserved between different proteins -- even ones whose function is very different. Sections of the mRNA will be excised, so the strand will be shortened and some of the information transcribed from the DNA will not be used.

I don't understand this at all. Why does it need to be translated? Isn't RNA/DNA a protein anyway? And what does "translated" mean? And how does this "translation" take place? Again, there is a chemical reaction taking place, which means that it is energetically favorable to store information in the protein, right? And if this is the case, why doesn't the protein just copy the DNA directly?
RNA/DNA are nucleic acids, not proteins. "Translated" means that the information stored in the mRNA (which was copied from the DNA) is taken and turned into a protein strand. The sequence is read in groups of three bases, called codons (see RNA Codon table), which correspond with a particular amino acid. Energy in the form of ATP is used to drive a lot of these reactions, so they are not necessarily energetically favourable.

This tells me that genetic information of less than three base pairs long is irrelevant. That is, if there are three sequences that all give the same amino acid, why do we need three sequences? Given that nature typically doesn't do irrelevant things, why should this be the case?
Mathematics -- there are more possible three base combinations of the four available bases than there are amino acids (at least the ones that are used in proteins, which is approximately 20). If some amino acids didn't have multiple codons that coded for it, we would have codons that did nothing which isn't an efficient use of energy. This will also affect the amino acid make-up, since amino acids with more codons will generally have a higher probablilty of making it into the protein -- this is chemically significant since this is largely what determines protein shape. Shape confers functionality.

This seems unlikely. How do all of these different molecules come in to play? And how does this happen in a finite time---naively there are 100! different ways for these different molecules to float by. So what gives?
It seems highly unlikely, yes. They interact the way they do based on chemistry. The details on some of the interactions may be less well known than others, but each individual reaction would happen on its own if we removed other substituents.
 
I'd like to say that biology is so full of exceptions, it's hard to find rules. But before listing every single exception, I thought we could at least give him a general idea of how stuff is working, some of the time. That way he could actually understand how the exception is an exception.

While I basically do agree I have to ask what precisely would be the exception and what is the rule (to be precise, there are of course no exceptions per se)? More to the point, a number of things listed in the various post are very specific to a certain system (e.g. 5' caps or polyA-tails).
 
While I basically do agree I have to ask what precisely would be the exception and what is the rule (to be precise, there are of course no exceptions per se)?

Oh, absolutely. I guess exceptions are things we discover working differently of our model organisms, eh?

More to the point, a number of things listed in the various post are very specific to a certain system (e.g. 5' caps or polyA-tails).

I guess I should have said "except when it doesn't" after everything I posted.
Which reminds me- Ben, would it help more if we gave you specific examples of mechanisms, or more general ideas? The more specific the example, the less likely it will be true for everything.
 
Last edited:
Ben:

You should get a good textbook on Cellular Function; another one on Introductory Zoology; and another one on Introductory Botany. They usually cover both extant species, as well as extinct species [the way things were eons ago] from the fossil record. That will give you a fairly good breadth understanding. You can then start filling in the blanks from there.

Glad no one thought I posted incorrect information. I hate it when I make typos or state something incorrectly.

When I double-majored, taking both physics classes and biology classes concurrently, it was interesting to see the biology majors struggle with physics [like solving a RLC circuit using differential equations] while the physics majors couldn't get a grasp on the complexity of biology. I guess things don't change much, do they?
 
Roman, well yeah. I work with pro- as well as eukaryotes. Thus it is hard for me to see one to be the rule and the other to be the exception.

Walter, to be honest I did not really read all your post. Do you really want me to comment on it? ;P

Ben, I would recommend that you simply concentrate on molecular biology and cell biology. The rest is likely to be of little interest to you at this point. And arguably it might already cover more than what most biophysicists need to know.
 
You realise that a cell, even a bacterial cell, is a computer, right?
It's a self-contained processor, a very complex one, but it interacts biochemically to external chemical gradients, which are the substrate, and the 'algorithm' is the DNA and RNA that every cell transforms into proteins.

But every process looks algorithmic - or has input and output type connections that channel stuff.

Every biochemical pathway you learn about has input molecules and output molecules. Nowadays we're starting to use things like DNA codes to realise molecular-scale computation - because DNA is a program for a cellular processor.
PENG YIN (1) ; TURBERFIELD Andrew J. (2) ; SAHU Sudheer (1) ; REIF John H. (1) ;
Affiliation(s) du ou des auteurs / Author(s) Affiliation(s)
(1) Department of Computer Science, Duke University, Box 90129, Durham, NC 27708-0129, ETATS-UNIS
(2) University of Oxford, Department of Physics, Clarendon Laboratory, Parks Road, Oxford OX 1 3PU, ROYAUME-UNI
Abstract

Intelligent nanomechanical devices that operate in an autonomous fashion are of great theoretical and practical interest. Recent successes in building large scale DNA nano-structures, in constructing DNA mechanical devices, and in DNA computing provide a solid foundation for the next step forward: designing autonomous DNA mechanical devices capable of arbitrarily complex behavior.

One prototype system towards this goal can be an autonomous DNA mechanical device capable of universal computation, by mimicking the operation of a universal Turing machine.
Building on our prior theoretical design and prototype experimental construction of an autonomous unidirectional DNA walking device moving along a linear track, we present here the design of a nanomechanical DNA device that autonomously mimics the operation of a 2-state 5-color universal Turing machine.

Our autonomous nanomechanical device, called an Autonomous DNA Turing Machine (ADTM), is thus capable of universal computation and hence complex translational motion, which we define as universal translational motion.

Like I keep saying.
 
Last edited:
Ben, how familiar are you with chemistry? What's the highest level chemistry course you've taken?

Well, I DO have a Bachelor's degree in Chemistry :) But I also took a graduate level Quantum Chemistry class to get out of taking biochemistry.
 
He's off somewhere, tugging away on some thing or other I'd say.

Nope:) Just banning trolls and taking pictures of whales. If anyone wants to see some really awesome closeups of humpbacks from my weekend furlough in Monterey Bay, PM me. Don't be disappointed if I don't send you a link, though, because the pics are on my personal blog.

Ok, I have a lot to digest, so I thank everyone in advance for their patience :)
 
Well, I DO have a Bachelor's degree in Chemistry :) But I also took a graduate level Quantum Chemistry class to get out of taking biochemistry.

:eek:
Sorry.
So you know plenty about chemistry and chemistry terms. That's good. It's just when you started calling RNA & DNA protein, I was sort of ???, but I guess that's more of a biochem thing.
 
I am probably at approximately Ben's level. I crudely understand how the information of DNA changes to slightly different code and structure in RNA and is utilized, three bases at a time (codons) to make small set of amino acids and they a potentially much larger set of proteins. Even that the proteins normally twist up into complicated 3D shapes (and sometimes they or a very similar shape) can "fit" or bind in a surface "well” of the cell wall to modify the activity of the cell - I.e. are used as control signals. What seem strange to me is:

(1) Why are there not more (than 20?) amino acids made? (4 letter code in groups of three to a "word" is 64 words, I think. Are some groups of three not chemically bound into a set of three?)
(2) How these amino acids join up to form proteins (if that is what proteins are) that the cell needs (and not "toxic trash" like snakes can inject into you)?
(3) How these protein molecules join up to make larger, well defined, structures like the 26 chromosomes? (and not just a big mess of "glue balls")

I.e. do humans understand what is guiding the production of and controlling the USE of proteins? (I do not expect a frog to jump out of a jar containing a solution rich in all the amino acids after vigoriously shaking it - something is missing.)
 
Last edited by a moderator:
(1) Why are there not more (than 20?) amino acids made? (4 letter code in groups of three to a "word" is 64 words, I think. Are some groups of three not chemically bound into a set of three?)

There is a redundancy in the genetic code. Three codons are stop codons, that is, they do not code an amino acid, but terminate the translation. With the exception of tryptophane and methionine (which is also the start codon) the amino acids are coded by more than one codon. There are variations of the genetic code, however only between very separate species.

(2) How these amino acids join up to form proteins (if that is what proteins are) that the cell needs (and not "toxic trash" like snakes can inject into you)?
The process of protein forming, the protein-biosynthesis, occurs in the ribosomes in a specific way by the formation of peptide bonds. The sequence of the amino acids to be connected is defined by the respective mRNAs. The tRNAs define the genetic code.

(3) How these protein molecules join up to make larger, well defined, structures like the 26 chromosomes? (and not just a big mess of "glue balls")
Chromosomes (at least the more sophisticated structures found in eukaryotes that you are thinking of) are formed by the interaction of DNA with histones and other DNA binding (and guiding) proteins. The basic units are nucleosomes, histones wound with DNA.
It is a rather complicated highly organized system that I am probably not able to convey in few clear sentences (at least not with gross, erroneous simplifcations). Especially given the fact that not everything is known in detail about the dynamics of chromosome organization yet.

I.e. do humans understand what is guiding the production of and controlling the USE of proteins? (I do not expect a frog to jump out of a jar containing a solution rich in all the amino acids after vigoriously shaking it - something is missing.)
Partially yes, but not for every (or even the majority of) protein(s). The production is pretty straightforward. Also there is a lot of knowledge regarding the signals that are involved in transporting the respective proteins to their destinations. But after that the diversity of protein functions makes it almost important to find generalized rules. Basically for a limited amount of systems the assembly functional protein complexes (if applicable) and their functions are known. Quite often also the assisting proteins (like e.g. chaperones) that are involved in protein folding and thus important for the finalized structure are also investigated.
Much more effort is necessary to analyze more complex protein-protein (or protein-whatever) interactions, as well as the basic functional characterization of novel proteins.

My favorite example (which my students probably already heard around a gazillion times) is E. coli. This bacterium is the genetically best characterized organism on earth, its sequence has been around for ages, however despite all the work of decades around 25% of its genome (and thus protein content, as prokaryotic genomes are packed tighter) has no known function.
 
Thanks. My question (1) is basically answered.
...The process of protein forming, the protein-biosynthesis, occurs in the ribosomes in a specific way by the formation of peptide bonds. The sequence of the amino acids to be connected is defined by the respective mRNAs. The tRNAs define the genetic code.
I think that the "m" of mRNA is for "messenger" and also that mRNA is a reasonably well defined sting of about 20 to 25 bases long with again some "end flags" part of the string to mark the start and end. (is that correct?) If the information for making a particular protein is in the tRNA, which I assume is a short section of the information in the long DNA, rewritten with one of the encoding letters changed, does the mRNA get made from the tRNA, very much like the long /RNA is made from the long DNA -I.e. sort formed up on it as the needed letters floating around in the "soup" stick to the pattern and then when fully formed "peal off" to be the mRNA?

Now can a patricular mRNA use any ribosomes (as the factory to build the protein it codes for or must it find an equally specailzed ribosome that makes only that (or a few other) proteins? As there are many proteins, I would guess that there are far fewer ribosomes so each must be quite a flexible factory - sort of like a quadradigitally control automatic factory. Perhaps all ribosomes are identical?


...Chromosomes (at least the more sophisticated structures found in eukaryotes that you are thinking of) are formed by the interaction of DNA with histones and other DNA binding (and guiding) proteins. The basic units are nucleosomes, histones wound with DNA. ...there is a lot of knowledge regarding the signals that are involved in transporting the respective proteins to their destinations. But after that the diversity of protein functions makes it almost important to find generalized rules. ... Quite often also the assisting proteins (like e.g. chaperones) that are involved in protein folding and thus important for the finalized structure are also investigated. ...
So, if I follow you, DNA is not just the coded information, but also part of the building up process to make more complex structures from proteins? (Part of your text I made bold) What are "histones" ? Are they a class of structures made of proteins? (that then have the DNA would on them to help it do its second function of building the complex groups of proteins into structures with functions)?
 
I think that the "m" of mRNA is for "messenger" and also that mRNA is a reasonably well defined sting of about 20 to 25 bases long with again some "end flags" part of the string to mark the start and end. (is that correct?)

Not quite. The mRNA contains the sequence information for a particular protein. Thus a maturated mRNA contains roughly 1000 to 3000 base pairs (bps) which code the protein (on average, this part can be longer or shorter). In addition there are shorter non-coding sequences at the beginning and end (including e.g. the ribsome binding site).

If the information for making a particular protein is in the tRNA, which I assume is a short section of the information in the long DNA, rewritten with one of the encoding letters changed, does the mRNA get made from the tRNA, very much like the long /RNA is made from the long DNA -I.e. sort formed up on it as the needed letters floating around in the "soup" stick to the pattern and then when fully formed "peal off" to be the mRNA?
Here we got a misunderstanding. The information is more on the mRNA side. That is, the mRNA encodes a particular sequence of amino acids and is necessary to create the protein. However how a codon is translated, that is, which amino acid will be built according to a specific codon is determined by the tRNA. It works that way: a tRNA possess a anticodon part in its sequence that is able to pair with a specific codon on the mRNA. Let us assume that the codon in question in AUG on the mRNA, and the anticodon on the tRNA would then be TAC. This TAC-containing tRNA gets loaded with a specific amino acid (in this case methionine). This means that during protein synthesis in the ribosome at some point the AUG of the mRNA would be read, the tRNA with TAC-methionine would be able to bind to this area and the loaded methionine would be connected to the already existing peptide being synthesized.
If by mutations or whatever other reason a tRNA with the anti-codon TAC would be loaded with, say, threonine, then this amino acid would be built in instead of methionine.
In other words, the mRNA got the sequence information of the protein, but the loading of tRNAs with the respective amino acid determines how this sequence is decoded in terms of amino acids.

Now can a patricular mRNA use any ribosomes (as the factory to build the protein it codes for or must it find an equally specailzed ribosome that makes only that (or a few other) proteins? As there are many proteins, I would guess that there are far fewer ribosomes so each must be quite a flexible factory - sort of like a quadradigitally control automatic factory. Perhaps all ribosomes are identical?
Ribosomes are universal protein factories. They only need the particular mRNA as blueprint to create different proteins.

So, if I follow you, DNA is not just the coded information, but also part of the building up process to make more complex structures from proteins? (Part of your text I made bold) What are "histones" ? Are they a class of structures made of proteins? (that then have the DNA would on them to help it do its second function of building the complex groups of proteins into structures with functions)?
Well, that are different questions, actually. The function of DNA is providing information, however the structure of chromsomes is also determined by DNA. Nonetheless I would not call it a structural function per se. Chromosomes are just a particular way to organize the DNA. Histones are specific species of DNA binding proteins.
 
I think that the "m" of mRNA is for "messenger" and also that mRNA is a reasonably well defined sting of about 20 to 25 bases long with again some "end flags" part of the string to mark the start and end. (is that correct?)

20 to 25 bases is 7 to 8 codons, which make 7 to 8 amino acids. These would be very small proteins, considering that you also need a start and stop codon.

If the information for making a particular protein is in the tRNA,

It's not. CharonZ was a bit confusing about that part. tRNAs recognize a codon on the mRNA-ribosome complex, and complementarily base-pair with the exposed piece of mRNA. This causes a conformational change in the complex, the tRNA releases the amino acid attached to it, and this amino acid is linked to the growing chain of polypeptides. I may be making some "gross simplifications", but that's essentially how it works, most of the time, in most organisms.

which I assume is a short section of the information in the long DNA, rewritten with one of the encoding letters changed, does the mRNA get made from the tRNA, very much like the long /RNA is made from the long DNA -I.e. sort formed up on it as the needed letters floating around in the "soup" stick to the pattern and then when fully formed "peal off" to be the mRNA?

mRNA is synthesized directly from the DNA, and often times, mRNA IS long, especially when it has to encode subunits for macromolecules.
Here's the wikipage on translation, or making proteins from the information on mRNA:

http://en.wikipedia.org/wiki/Translation_(biology)

Now can a patricular mRNA use any ribosomes (as the factory to build the protein it codes for or must it find an equally specailzed ribosome that makes only that (or a few other) proteins? As there are many proteins, I would guess that there are far fewer ribosomes so each must be quite a flexible factory - sort of like a quadradigitally control automatic factory. Perhaps all ribosomes are identical?

I can't speak to how different ribosomes can be, but as for their similarities, you can take the DNA for genes coding for various proteins from one species, put it in another, and get that protein. Bacteria have been used to synthesize the insulin in diabetes medication, for instance. Green Fluorescent Protein, or GFP, was found to be the protein that makes jellyfish glow in the dark. The DNA for this protein has been successfully integrated into the chromosomal DNA of everything from fish to monkeys, which means we get things like glow in the dark bunnies. This means that a protein that evolved in jellyfish is being created by the ribosomes in a higher vertebrate.

So, if I follow you, DNA is not just the coded information, but also part of the building up process to make more complex structures from proteins?

Yes, but not in the way you are thinking. I'm unaware of any cases where DNA is directly involved in protein synthesis. Proteins interact with DNA all the time. For instance, several proteins must come together on DNA to begin making RNA copies of it. These proteins do things like melt the DNA, increase specificity, or increase efficiency of transcription (making RNA copies).

DNA pretty much just sits there and gets fiddled with. Chaperon molecules and changes in physiological conditions (pH, Calcium ions, salt concentrations, etc) are largely what affect the proteins in forming their 'functional' structure. Not that DNA binding doesn't also change their structure, it's just that in order to be a molecule for binding DNA, they first need to be folded the right way, so perhaps a calcium ion needs to bind, or it needs to be de-phosphorylated.

(Part of your text I made bold) What are "histones" ? Are they a class of structures made of proteins? (that then have the DNA would on them to help it do its second function of building the complex groups of proteins into structures with functions)?

Histones are proteins that DNA gets wound around. They function in turning genes on & off, by making different parts of DNA available, as well as keeping the DNA compacted and safe.

Many, many genes in most of our cells right now are not active, some of which are due to the way the DNA is packaged with the histones. These histones can be modified to either increase or decrease the chance that that particular part of DNA will become active. Methylation is one of these ways. It's hypothesized that methylation of somatic cells (such as our cheek cells or the udder cell of a sheep) is what makes cloning so difficult. Since a great deal of genes are turned off due to methylation patterns_ what parts of DNA are being blocked or not by histones, there's no guarantee that they will turn back on in an embryo.
 
Back
Top