Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Primary structure

  1. Sep 25, 2012 #1

    what are the available methods to determine the sequence (primary structure) of a protein?

    from where one can start (what info he can gather) and how he can deduce the primary structure?

  2. jcsd
  3. Sep 25, 2012 #2
  4. Sep 25, 2012 #3
    sorry, not even close to what I am asking
  5. Sep 25, 2012 #4
    Perhaps if you could explain a bit more in detail what you need, it could be helpful.

    The previously mentioned site notes that the advent of large amounts of genomic information has made it possible to predict protein sequences based on the corresponding nucleotie sequence. Also, mass spectrometry has been tremendously helpful in this area - see, for example, this perspective/review article.
  6. Sep 25, 2012 #5
    if I give you the volume of the peptide (let's only talk about small peptides), the charge/polarity/hydrophobicity of the various parts of its surface, its precise dimensions, will you be able to predict its sequence?

    if not, why?
    if not, is there anything else you would need to do it?
  7. Sep 25, 2012 #6
    If we say that there are 20 common amino acids, and only 20, and that they always link in the same way (e.g. no gamma-gln linkages) then there are 3.2 million pentapeptides. There is only one possible way that you can use the physical and/or chemical properties to identify any one of those pentapeptides in an intact molecule: prepare a suitable crystal of the material, and determine its crystal structure by X-ray diffraction (minimum of a month of research if you are familiar with the techniques and fortunate enough to find a way to prepare the crystal). Otherwise, destructive techniques are always necessary. Obviously those that require only a small amount of material are preferable. The Edman degradation does lend itself to automation.

    Sequencing methods that are used in practice are just exactly those that are described in the U Wisconsin article linked above.
  8. Sep 25, 2012 #7
    I dont have the protein or I cannot crystalize it

    is there a workaround?

    can I "generate computationally" a crystal, given its physical/chemical properties (which is what I only know) and then computationally again hit it with x-ray and determine its sequence?
  9. Sep 25, 2012 #8
    OK, the way your original post seems to read is "Suppose I have here a sample of a protein. How could I find out its sequence?"

    It is now sounding like you are asking something a little different.

    The X-ray diffraction technique depends on measuring the angular position and intensity of x-ray "spots" that you can find with a photon detector when you shine a beam of X-rays onto a small crystal of a substance.

    If you do not know the sequence of a peptide, there is no way of telling where these spots will be theoretically. In fact, to predict the positions of the diffraction spots you would also need to know the secondary structure, and the way that a large number of molecules could make a regular packing.

    If you do not have the peptide, there is no way that you could say accurately enough what any of its physical or chemical properties might be anyway.

    So are you trying to identify a small peptide that you have a sample of? Or design a small peptide to have a particular physical or chemical property? Or to work out what some of the physical or chemical properties of a particular small peptide with a known sequence?
  10. Sep 25, 2012 #9
    Uh, what? You asked about methods of determining primary structure. I gave you a link with three distinct methods of doing just that. If you're changing your question, you could have just clarified and said that instead of implying my reading comprehension sucks when your initial question was clear as crystal.
  11. Sep 26, 2012 #10
    no, since I dont have sample of the peptide (I just know some of its physical/chemical properties, and maybe some parts of its sequence or maybe that some specific aminoacids are not present in the sequence, etc)

    yes this is my quest basically:

    design [try to find possible/compatible sequences of] a small peptide to have a particular physical or chemical property

    I bet it's hard and not very accurate, but I would like to know the current abilities we have regarding this

    no, I know [some of] it's physical/chemical properties, its sequence is what I dont know

    ofcourse, if finding some specific physical/chemical properties will lead me to elucidate its sequence, it would be interesting
  12. Sep 26, 2012 #11
    no bad feelings, let's agree what I was asking was:
  13. Sep 27, 2012 #12
    Fair enough. It just seemed like a bit of a blunt response.
  14. Sep 27, 2012 #13


    User Avatar
    Science Advisor
    2017 Award

    Perhaps you're looking for something like this?
  15. Sep 27, 2012 #14
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook