Yo Pent, genes and biotech

[Nephew Twiddleton]

Also a convenient place for me to drop stuff specifically relating to genetics rather than biology as a whole, since I'm taking genetics in two weeks.

So, just as a quick run-down of how DNA and RNA work together before getting deep into the protein bits:

-DNA and RNA are strands of genetic instruction, the main difference between them being that the sugar molecules in their bases differ by one oxygen atom.
-Both use 4 nucleotides (bases), Adenine, Cytosine, and Guanine for both (differing only in that one oxygen atom, natch) and DNA uses Thymine where RNA uses Uracil. Cytosine always pairs with Guanine. Adenine always pairs with Thymine or Uracil. This is basically due to whether or not they can form 2 or 3 hydrogen bonds on the center of the double helix.
-Nucleotides are made up of a phosphate group, a sugar (ribose/deoxyribose/other- more on that in a bit), and a nitrogenous base. The phosphate group of one forms a phosphodiester bond with the sugar of the one next to it, this creates the strand, and nitrogenous bases, which determine what nucleotide it is, forms weaker (and therefore, easy to unzip) hydrogen bonds with the nitrogenous base of its counterpart.
-There are more than one kind of nucleotide, and more than one kind of nucleic acid. ATP, for example, which is the fuel our cellular processes, is basically a modified adenine molecule with three phosphate groups. Further, there are nucleotides that form XNA- nucleic acids that for some reason or another, life on Earth decided not to bother incorporating, but can be used it the fields of synthetic biology and biotechnology (creating exotic nucleic acids that won't interact with ours if it got loose, or even in capping off replication early in biotech). XNA sounds intriguing to me and I think I'll read about it more in my own time as well.
-DNA is an instruction manual, genes are particular tasks in the instruction manual, RNA is the instruction being read outloud, and proteins are the end result of the instructions.
-Triplet sequences of nucleotides code for one of twenty amino acids, or tell you to stop coding. There are 64 possible combinations for codons. Only one sequence codes for "Start" and "Methionine." SIX code for Leucine, and three code for "stop". I'm not exactly sure how these redundancies work, or why Methionine gets no redundancy. Most get 2 or 3. The redundancies help mitigate the chance of damaging mutation. If a syllable can be coded for in a couple of different ways, spelling errors don't have to be that nonsensical.
-DNA unzips, messenger RNA gets grafted on to the exposed nitrogenous bases, and makes a mirror copy. Since most DNA doesn't actually code for anything, the mRNA needs to get edited. The edited mRNA then gets paired with transfer RNA. tRNA is what ends up going on to synthesizing amino acid sequences, which then fold around and create 3 dimensional proteins.

I'll address some of your questions in the next post.

[Nephew Twiddleton]

Got an A in Microbiology, and a B- in general bio. I'm expecting probably a C of some sort in biotech and an F in algebra due to inability to catch up on the course work. So I'm just going to take the placement test next week and make the course unnecessary.

Has anyone figured out dna's loop syntax yet or anything equivalent? Can't find any info on how the shit actually works. I'm wondering how developed this biotech thing has gotten and I'm not hooked into any good feeds, other than end results-based bullshit.

It's the order that amino acids are coded for if I understand your question. Each amino acid is coded for by just three nucleotides, and there's only twenty of them (other triplets function as periods in the sentence, and each sequence always starts with methionine.) Each protein has four structural levels, each building on the previous, and the first is determined by amino acids sequences.

I love your terminology - "periods in a sentence" Perfect! How many words do we know? Do we know all of them? These protein layers are a kind of branching logic that builds complexity from a less complex instruction?

Where's the best place for an idiots guide to this shit, where I can see it going from how the dna code creates the instructions to form cells, what type, how many to make, structural formulae and shit and how this information unfolds from the program. I have a vague idea of a couple of the mechanisms but is there any kind of broad-brush, a-b-c thing you know of?

Self replication is the biggest mindfuck since object oriented from where I'm sitting (I had a hard time with OOP, back at the start) but this shit is a whole new level. There's instructions in the dna code that make the machinery that carries out the instructions in the code, right? How the fuck do you code for that? It's like some kinda weird organic unzipping algorithm. 

I'll see if I can poke around for some stuff when I get home.

How many words we know:
Good question, and I'm not exactly sure. If there's not an infinite variety of proteins out there, there may as well be, so there's always more for us to crack. Consider that we discovered DNA wasn't just a random, apparently useless biochemical until about 70 years after it was discovered, and we didn't even know what shape it was until about 20 years after that. Or that our own genome has only been mapped for what... 10 years? 15? And consider the variety of life and viruses, and that we're only aware of about 10% of all of it (and that's not even the genetics of them. That's just number of species), and those are the ones that are still around. It's a really robust language for one that only has 20 syllables!

Proteins:
The best way to think of how they work is a key that fits a lock. "Lock and key" is arguably the most cliched phrase in biology. A protein only does what it does because it binds in a specific way to other proteins and enzymes.

What they say, how they say it, and when they say it:
That's a pretty complex question, and probably one I can't adequately answer at this point. But, it's thought that all that extra DNA that doesn't code for anything are the instructions there. Basically, the context for when a gene codes for a protein, or if it codes in conjunction with other genes.

Self-replication:
It fucks with my head too, not the least reason being that DNA codes for RNA, RNA codes for proteins and enzymes, and proteins and enzymes are needed to unzip DNA and start that grafting of DNA or RNA nucleotides onto the strand. You basically need one for the other to happen in the first place, which is why the origin of life is hard to figure out.  Ok, now I'll see if I can find some clips or something.

[Nephew Twiddleton]

This is basic run down of DNA to protein, with some better explanation of how primary to quartenary structure of proteins happen.
https://www.youtube.com/watch?v=itsb2SqR-R0

These are a couple of lectures on genetics that I found interesting. It's an hour and a half long each and it might be a bit meaty for your purposes, but, it shows how the basic run down above isn't the whole story.
https://www.youtube.com/watch?v=_dRXA1_e30o
https://www.youtube.com/watch?v=dFILgg9_hrU
https://www.youtube.com/watch?v=e0WZx7lUOrY

[P3nT4gR4m]

Thanks twid. Can't watch the clips til later but hopefully they'll put me on the right track. I'm looking at DNA as my next coding platform. I'm expecting I'll be doing it through a much more abstracted interface but, just like with my current silicone platform, I suspect it will be helpful to have a rough idea of what's going on with all the little chips and transistors and buses, without necessarily going into the depth a cpu-designer would.

The DNA-RNA-Protein thing is how I thought things were working but I wasn't sure if proteins were the "robots" or if they did something bigger so thanks for that. The four layer thing is awesome!

[Nephew Twiddleton]

Thanks twid. Can't watch the clips til later but hopefully they'll put me on the right track. I'm looking at DNA as my next coding platform. I'm expecting I'll be doing it through a much more abstracted interface but, just like with my current silicone platform, I suspect it will be helpful to have a rough idea of what's going on with all the little chips and transistors and buses, without necessarily going into the depth a cpu-designer would.

The DNA-RNA-Protein thing is how I thought things were working but I wasn't sure if proteins were the "robots" or if they did something bigger so thanks for that. The four layer thing is awesome!

No probs, man. This is going to help me too, at least to get to thinking.

Regardless of what I do, it's probably going to form the basis of what I'm going to be doing for my career, even if I go into pure academic research- the hypothesis I have will require me to look at a lot of genetic code, and actually thinking about it in terms of amino acids offer me the chance to do it at least three times as quick, which is something I didn't realize until rereading the thread now. Or even starting at the protein level and then working my way down.

[Nephew Twiddleton]

Some links on XNA (Xeno Nucleic Acids, incidentally) that I haven't looked at yet:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2909387/
http://io9.com/5903221/meet-xna-the-first-synthetic-dna-that-evolves-like-the-real-thing
http://books.google.com/books?id=lr0SgX1Hx2gC&pg=PA151#v=onepage&q&f=false
https://www.landesbioscience.com/journals/artificialdna/
http://www.popularmechanics.com/science/health/genetics/xna-synthetic-dna-that-can-evolve-8210483
http://news.nationalgeographic.com/news/2012/04/120419-xna-synthetic-dna-evolution-genetics-life-science/

[Nephew Twiddleton]

Ok, so, we can do computer models of what kind of protein an mRNA will produce, just based on what sequences form the protein. In otherwords, we can run a nucleotide sequence through a computer and the computer might say, "This is bullshit. Can't exist." One of the things to note about proteins is that they have a primary structure (the resulting strand of RNA from instruction), secondary structure which is determined by hydrogen bonds and their resulting alpha helices or beta pleats

(Digression that might be illustrative- there are different kinds of chemical bonds. Ionic bonds are where one atom donates an electron to another, and they hang out due to the resulting magnetism. Table salt is an example of this. Covalent bonds are those in which the electrons are shared between atoms because of valence electron shells. For example, the first shell demands two electrons. Hydrogen has one proton and one electron and one empty spot in the lowest valence shell. So it would naturally form a covalent bond with another hydrogen, so that both have a complete shell between them. That's the first shell. The second shell demands 8 electrons, which is why lithium does violent shit in its pure form, or why oxygen allows for flammability. Oxygen has 8 protons, but loves 10 electrons. Hence, water. Each hydrogen gets two electrons, and the oxygen gets two electrons. Best threesome ever. This brings us to the hydrogen bond. A water molecule looks like Mickey Mouse. The ears have a slight positive charge because the electrons like hanging around the oxygen, and the chin has a slight negative charge for the same reason. So the ear of one, likes to touch the chin of the other, due to mild polarity.)

I'm going to leave off with that digression for the moment because I've had a bit of wine due to holiday tomorrow, and I'm likely to ramble at this point. But the types of bonds are important in understanding how proteins work.

[P3nT4gR4m]

Okay the digression was enlightening but I'm thinking TMI from my point of view The proteins thing, tho is totally the bit that's fucking me up. I suspect it's the quintessential difference between coding for abstract information (my current line of work) and coding for self-replicating smart matter (where bioinformatics is headed) I'm thinking the crux of the matter is that coding computer uses instructions to manipulate data whereas coding bio uses little robots to build stuff?

I'm trying to come up with an analogy for proteins and I'm fucked if I can. If I'm reading this right, proteins are an intermediate step between DNA/RNA "operation" and cells although sometimes they can be the end in itself. How does one get from protein synthesis to a cell or an organism?

[Nephew Twiddleton]

Okay the digression was enlightening but I'm thinking TMI from my point of view The proteins thing, tho is totally the bit that's fucking me up. I suspect it's the quintessential difference between coding for abstract information (my current line of work) and coding for self-replicating smart matter (where bioinformatics is headed) I'm thinking the crux of the matter is that coding computer uses instructions to manipulate data whereas coding bio uses little robots to build stuff?

I'm trying to come up with an analogy for proteins and I'm fucked if I can. If I'm reading this right, proteins are an intermediate step between DNA/RNA "operation" and cells although sometimes they can be the end in itself. How does one get from protein synthesis to a cell or an organism?

RNA is the intermediary. DNA is one end and proteins are the other. And, here's the thing that I got distracted from the other night, is that the important thing about proteins is the shape of the protein, and how it can change shape if it is allosteric (has two shapes depending on what binds with it). So you have all those various chemical bonds which are important to know, and you have primary structure which is purely sequence, secondary structure which starts to create alpha helices and beta pleats (alpha helices make good transmembrane channels. If you want stuff to go in and out of the cell freely, alpha helix) tertiary structure is where a protein pulls itself into a three dimensional form, and quarternary structure is where two or more proteins pull together into a different protein. Human hemoglobin is a prime example. It consists of two groups of two proteins massed into one hemoglobin.

So, the insight that I gained this semester is that it's not the sequence, but rather the shape that the sequence makes. Lock and key is a cliche for a reason. The three dimensional structure, which does not change so long as the sequence of amino acids is unchanged, is really what it's all about.

Incidentally, the hydrogen bonds are important because they're weak enough to unzip DNA, but, again, because of the three dimensional structure of proteins and the sheer number of them in protein structure give a stable macromolecule. Unless you heat it up until it collapses entirely.

[P3nT4gR4m]

Not really genes and biotech but kind of genes and biotech.

[Reginald Ret]

Twid, I love your explanations, I think we could work together on a kids guide to biochemistry if I ever get my act/shit/mind together.

P3nt, I love your questions, you ask questions I wouldn't think existed yet when you do they make a lot of sense.

[P3nT4gR4m]

This seems like big news

[S]

Considering how frigging long it takes to crystallize a protein that news is extremely big. Too bad XFELs ard so expensive...

[Mesozoic Mister Nigel]

High-definition XRC. That's cool.

[P3nT4gR4m]

Also heard about a camera that shoots 100 billion frames per sec. Dunno about scientific shit but if I could shoot helmet cam at even a tiny fraction of that, it'd be king of awesome!