Question

1. For each polypeptide sequence listed, choose from the options given below to indicate which
secondary structure the sequence is most likely to form upon folding. The nonpolar amino acids
are italicized. Explain your choice(s).

A. Leu-Gly-Val-Leu-Ser-Leu-Phe-Ser-Gly-Leu-Met-Trp-Phe-Phe-Trp-Ile
B. Leu-Leu-Gln-Ser-Ile-Ala-Ser-Val-Leu-Gln-Ser-Leu-Leu-Cys-Ala-Ile
C. Thr-Leu-Asn-Ile-Ser-Phe-Gln-Met-Glu-Leu-Asp-Val-Ser-Ile-Arg-Trp

amphipathic α helix hydrophilic α helix
hydrophobic α helix amphipathic β sheet
hydrophilic β sheet

Answer 1

Hm. Something seems a bit fishy about this question.. o.o

I don't think, though I'm not certain, that given such a short list of amino acids, you can tell, by eye, what secondary structure will be formed. Predicting protein structure is a huge task requiring very complex algorithms, and is still being worked on. Maybe @blues can shed some more light on this.

Answer 2

This is very interesting. Looks like there are quite a number of algorithms and programs developed to tackle this program.

For example, you may want to look at the GOR method: http://en.wikipedia.org/wiki/GOR_method

This lets you paste in your sequence and applies the algorithm: http://cib.cf.ocha.ac.jp/bitool/GOR/

Oh and this little script converts your three-letter code to the more common one-letter abbreviation, used above^: http://molbiol.ru/eng/scripts/01_17.html

Answer 3

People do make shrewd guesses about secondary structure based on primary structure. These guesses are of dubious use. I have seen informal, in-house studies conducted on very sharp protein biochemists and molecular modelling types in which the people with the big PhDs successfully infer the right secondary structure ~20% of the time. Which leads me to wonder whether the expert-designed computer programs might not bat that much better.

That said, there are situations in which a guess - even a wrong guess - is useful as a starting point in homology modelling or some sort of iterative, refinement based process.

As some guiding points, successive residues in a beta sheet stick out on opposite sides of the sheet. So if you get a sequence with alternating qualities - for example, alternating residues are hydrophobic and hydrophilic - then you're looking at an amphiathic beta sheet.

Or if you scout out a heptad repeat, or if you have salt bridge - hydrophobic - hydrophobic - salt bridge residues, the first raises the possibility of an alpha helix while the second is the common sequence basis of two alpha helices coiled around each other.

At this point it is dark magic and guessing. For whomever comes up with a proper solution to the protein folding problem, that particular Nobel is still out there for the taking...