Time for the glass eye test to be inserted into CASP

What’s CASP ? — it stands for Critical Assesments of Techniques for protein Structure Prediction. An amino acid sequence of a protein is given to the people who try to deduce protein structure (e.g. the disposition of all the atoms of the protein in 3 dimensions) from just the amino acid sequence. The sequence (or sequences) are distributed to all, then various groups fire up their algorithms and computers and come up with a structure which they submit to CASP. “Targets are chosen from among those proteins whose structures are expected to be solved shortly, mainly by one of the structural genomics centers.” (WikiPedia). CASP uses variety of measures to see how close the submitted 3 dimensional structures are to the solved structure.

What’s not to like? Well for one thing, the groups entering the fray know that such a structure exists. They know they should find something. But not all proteins have a single defined structure. Consider ubiquitin. [ Science vol. 320 pp. 1428 – 1429, 1471 – 1475 ’08 ] Ubiquitin bound to various proteins has been studied by Xray crystallography — some 46 different shapes have been found. Using solution NMR and a trick known as RDC (residual dipolar coupling) and molecular dynamics simulation (now up to all of a microSecond), the 46 different ubiquitin shapes have been found in solution.

For another thing, not all proteins HAVE a shape. [ Nature vol. 438 pp. 605 – 611 ’05 ] “More than 100 ‘natively unfolded’ or ‘intrinsically unstructured’ proteins have been identified; their evolutionary persistence strongly implies that they have an important function. MARCKS is a prototypical natively unfolded protein with a high fraction of negatively charged residues and a few hydrophobic residues; thus it exists in an extended conformation.” Moreover, consider this — It appears that a disproportionate number of proteins involved in cell signaling are intrinsically unstructured — which essentially amplifies the allosteric response. [ Proc. Natl. Acad. Sci. vol. 106 pp. 6887 – 6888, 6927 – 6932 ’09 ].

As for me, I find it quite miraculous that ANY protein has just a few shapes (although those making up living things must have them). More on this in future posts.

Also native proteins just aren’t that stable. Denaturation is the process of conformational change by which a biologically active protein loses the ability to perform its function. Proteins denature with energy input of .4 kiloJoules/amino acid. This means that a 100 amino acid protein has a binding energy of 40 kiloJoules/mole. The carbon carbon bond is 348 kiloJoules/mole and a hydrogen bond is 20 kiloJoules/mole. (Sorry, I can’t remember the source where I took this down but it’s pretty standard).

Now to switch gears a bit. One of the hardest things for a fledgling doc to learn, is how to do a decent physical examination (PE). Whole courses are given in the subject. One of the hardest parts of the PE is looking into the back of the eye with an ophthalmoscope. This is important (particularly in neurology) because it’s the only place in the body where you can actually see a nerve (the optic nerve) along with arteries and veins (everything else you do in the PE resuls in just inference about them, rather than observation). If the optic nerve appears swollen then you know that pressure inside the head is elevated (always serious).

The temptation to fudge is ever present. This is why the patient with the glass eye is so valuable. If one is present and the unlucky med student attempts to say that they saw the nerve, it’s time for a lecture (hopefully not too sadistically). It didn’t happen to me, but the basic lesson is to report what you see honestly, whether you understand it or not — there’s probably someone smarter or with more experience who will (that’s what I heard when I reported a bunch of findings that just didn’t make sense).

But they never will if the data they get is fudged. Sadly, there is a reasonable possibility at this point that exactly this has occurred at the University of East Anglia in their Climate Research Unit. Too early to tell, but the EMails I’ve read certainly smell (and they’ve not been denied by any participant as inauthentic or altered in any way). Just google Climategate for the sordid details. Or have at look at the EMails yourself at http://www.eastangliaemails.com/

So it’s time to give the CASP troops either a random sequence of amino acids, or a protein known to be unstructured (hence not crystallizable) and see what they come up with. Their work would have a good deal more credence if they didn’t come up with a structure.

Post a comment or leave a trackback: Trackback URL.


  • luysii  On January 4, 2010 at 11:41 am

    Is there any hope of predicting 3 dimensional protein structure from the 1 dimensional sequence of amino acids making it up. Consider the following:

    [ Proc. Natl. Acad. Sci. vol. 106 pp. 21011 – 21012 ’09 ] The L45Y mutation in G(A) a fragment of protein G of streptococcus which binds serum (??) changes it from a 3 helix bundle (L45) to a 4 stranded beta sheet (Y45). It’s barely a protein, being between 45 and 56 amino acids long (just how long is it?). 85% of residues change their secondary structure in the two forms

    “When I first heard this result in a public seminar, my mind literally began to reel, leaving me dizzy and slightly nauseated as all hope of understanding how sequence encodes structure seemed to suddenly vanish”

    The other side of the coin is that it is possible to recover stably folded proteins from mutagenized libraries to substitute 50% or more amino acids without changing the proein fold or greatly lowering the stability. Most such sites are found on the protein surface.

    Segments of 8 amino acids have been found which have different secondary structures in different tertiary structures (Proteins vol. 30 pp. 228 – 231 ’09). They are called chameleon sequences. Chameleon sequences ‘usually’ consist of both strong helix-forming and strong beta strand forming amino acids.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: