how much knowledge we have of the human dna

[bwilly]

Hello,

since often this topic is discussed in terms of percentage, and science writers

speaks with very different terms on that subject , I would like to know at what

point knowledge of the dna is and what are next milestones (and obstacles)

to understand fullfill how dna works and, in particular, how human dna works.

Thanks,

wily

[Ringer]

Both what we know and what we don't are very long lists, could you perhaps be more specific in what you are looking for? Also, DNA doesn't, for the most part, work much differently in other organisms when compared to humans.

[CharonY]

Actually I am not aware of any kind of uniqueness of human DNA compared to other animals, for instance. The major issues in the use of DNA information is much in terms how that information in conjunction with the cellular content, result in physiological effects and reactions.

DNA alone is just a molecule with a certain variation in its base composition.

[bwilly]

Both what we know and what we don't are very long lists, could you perhaps be more specific in what you are looking for? Also, DNA doesn't, for the most part, work much differently in other organisms when compared to humans.

 

Well, the question is general so answers could be general

As an example, seems that we know a lot about the part of dna that codifies (about 1.5 percent).

But what about the rest? At which point we are in understanding regulatory sequences and what

in general is called junk-dna? Which knowledge we have in the effect of changing a gene expression

and the effect it has on the phenotype? And know if changing a single gene let the dna in, how could

I say, "coherent" state.

[CharonY]

Well for the coding regions we can (with some uncertainty) predict the gene products, of which we only know a relatively small proportion the biochemical function (e.g. enzyme activity) and much less about their physiological role within the organism. This is basically true for most genomes, not only human. Junk-DNA is still a hot topic and we have at best from a small fraction functional information. The effect of quantitative changes is quite an ongoing topic in systems biological approaches, but in most cases we barely have qualitative information, much less quantitative one. The translation from genome functions and their quantitative interaction (the latter is really the crucial point) is barely in its infancy.

[alpha2cen]

The translation from genome functions and their quantitative interaction (the latter is really the crucial point) is barely in its infancy.

The research about small DNA organism is useful to understand the relation from dna code to genome function. And system biology is also useful too. There has been many research done about dna to gene.

But, we have not made real artificial life in the computer. Artificial life in the computer ==>Real organism at the experiment might be important in this area.

Edited August 26, 2012 by alpha2cen

[dmaiski]

some key milestones that i can think of in terms of DNA, genetic engineering, ect... would be:

 

efficient simulation of DNA-->tertiary protein (this can be done at the moment, but the algorithms are at the best of times only about 80% correct), this leads on to whole genome simulation of an organisms development(this is only a problem of computation)

 

most of the actual manual labour for decoding DNA and its analysis has already been done, at the moment the process is just being streamlined for efficiency, speed and volume.

 

most notably mass screening of whole genomes and the analysis can be good milestones; ie the first 1000, 10000, 100000, genomes screened would give insight into patterns and effects of certain genotypes when related to the observed phenotype, this would really be the most important component for genetic treatment of disease and various other desirable characteristics

 

The previous comment also ties into another component vital for genetics, public acceptance of genetics and mass utilisation of the treatments offered. this will influence investment and research into genetics and is the deciding factor of how far we progress into the field.

 

another milestone is then genetics become commercially viable, this has already happened with plants(GM crops, still not accepted in the EU) and animals(although pet shops still dont stock glow in the dark cats, they do stock glow in the dark fish thou) the next milestone in this progression would be "GM humans" ie. designer babies(this would require detailed understanding of the effects of genes and their interactions, and demonstratable advantages for the treatments)

 

 

 

what we currently know about genes(don’t quote me on this I'm just taking these numbers from memory and its a bit shoddy)

we have somewhere in the area of 100,000 protein structures mapped("As of Tuesday Aug 28, 2012 at 5 PM PDT there are 84223 Structures" thank you pdb)

we have mapped the human genome

we have some notion of how structure relates to function(but this is only based on what we have seen before and not a real understanding of proteins)

we know a lot about regulatory systems for DNA transcription and intron exon boundarys the use of micro RNA for various manipulation of transcripts and as protein regulators and various other details

all that we know comes down to experimental knowledge, most of which has not been compiled or analysed sufficiently to allow theoretical knowledge to translate to practice.

 

a nice analogy is we know that you can build a tower and we can build a wall, and indeed maybe even a small house, but we haven’t quite gotten to the stage of building sky scrapers.

 

 

 

 

 

If I got something wrong please don’t kill me, I'm scrawny and wont taste good in a stew...

 

 

 

 

[Jens]

We know a lot (on all the topics mentioned below) and a lot of things can be explained, but let me just try to summarize some of the issues. This might make the situation clearer:

1. Splicing: Talking about the 1.5 % of DNA that codes: Since we know the genetic code (= how DNA translates into amino acids) we can predict how the protein will look like from a chemical standpoint (the sequence of amino acids). However, even this sometimes does not work out since in the nucleus the messengar RNA is rearranged by cutting out some pieces ("splicing"). To my knowledge this cannot be predicted correctly in cases were the remaining parts are small (since you cannot see the open reading frame without stop codons). But in general this is a minor issue.
2. 3D Protein fold: Actually we cannot calculate the real 3D structure of proteins correctly. This is a huge issue, since all the catalytic or other function of a protein lies in its exact 3D structure and cannot be predicted by its sequence of amino acids. We are only getting there very slowly by creating huge databases of known structures. So if you have bad luck and the protein does not ressemble closely in amino acid sequence one with known 3D structure, you will not be able to predict it. This is why most structures are still determined by Xray cristallography. The issue with this is, that there is no systematic way to obtain crystals of proteins (might take years, or never work out).
3. Protein primary function: For proteins not related to proteins with known functions there is no systematic way to find out what they are good for. It might take decades. For example it is known that the Nef protein of HIV is not needed in cell culture but is essential for the survival of HIV in humans. So in some way it is probably disturbing the human immuno response. However, even after 25 years of research in the whole world it is still not really clear what the Nef protein really is doing (even though there is some progress). The only way which is working is the other way round: If you can measure the function of a protein, you can purify it, sequence it and look for the corresponding DNA, clone it and produce it in large amounts for further research. So point 2 and 3 together mean, that we are not really able to read the language of life. If a protein with function B has evolved out of a protein with function A, we cannot systematically see that it has now altered the function. You have to measure it (and it is hard to measure, if you do not know what you should look for).
4. Protein regulation: Nearly all proteins have in addition to their primary function also specific functionality to regulate the primary function (activate, inhibit). Our knowledge of these regulatory functionality is much less complete. And also here, you can never be sure that you have detected all of them (because there are often multiple). This is because we cannot predict correctly which smaller molecules will bind to a protein, even with a fairly good known 3D structure.
5. long regulatory RNAs: The vast majority of the "junk DNA" is actually transcribed (at low copy number) to RNA. We only know very few examples how this works (e.g. inactivation of the second X chromosome in females). For most of the rest we have no detailed knowledge at all. It is fair to assume that per length there is not so much information in the non-coding DNA than in the coding DNA, but since it is far more than 95% of the DNA, you can assume that there might be as much information and function in it as in the coding DNA. Remember: the biggest enzyme of all -- the ribosome (catalyzing protein production) -- is actually a catalytic RNA, which is just surrounded by some helper proteins. We are not at all at the end of understanding what RNA actually does in the nucleus.
6. Order in space in eukaryotes: Just a few months ago it was published that chromosomes in the nucleus actually have an order in space and form neighborhood patterns specific to the different human cell types. So it looks we are just starting to really understand what the nucleus is actually good for (take for example the puzzling fact that only cells with nucleus form complex multicellular life forms).
7. Even if all points from 1-6 are solved, that does not mean that you automatically understand how the complex system of a cell works altogether.
8. Even if all points from 1-7 are solved, that does not mean that you automatically understand how the complex system of a complete human body works together or how the human body is formed during embryogenesis.

So in summary: There are still a lot of things to do. We cannot really read the language of life. And of course we are even more far away to speak the language of life (we cannot design a new catalytic functionality from scratch on a computer).

Edited October 20, 2012 by Jens

[Bill Angel]

 

Well, the question is general so answers could be general [img=http://pub.scienceforums.net/public/style_emoticons/default/smile.gif]

As an example, seems that we know a lot about the part of dna that codifies (about 1.5 percent).

But what about the rest? At which point we are in understanding regulatory sequences and what

in general is called junk-dna? Which knowledge we have in the effect of changing a gene expression

and the effect it has on the phenotype? And know if changing a single gene let the dna in, how could

I say, "coherent" state.

Interesting work was published last year about progress understanding the function of so called junk DNA:

Bits of Mystery DNA, Far From Junk, Play Crucial Role