Hydrophobic Residue Fraction Calculator

This percentage is a mole fraction of residues, exactly like every other composition figure on the site. It is not a mass fraction, and the two are converted the long way round: each residue count has to be weighted by the tabulated mass of that residue before the two totals can be compared. The average molecular weight calculator performs that weighting and is the right place to go when the question is about mass rather than about count.

It is also not a surface fraction, which is the more tempting confusion, because surface exposure is what actually governs interaction with water. A count cannot distinguish a residue on the outside of a folded structure from one buried in the middle of it, and it cannot distinguish either in a chain that is not folded at all. The number describes the chain the sequence describes, which is a list of residues in an order. Everything about how those residues stand in three dimensions is absent.

The hydrophobic class here includes the three aromatic residues, which are also counted by the aromaticity calculator . That double counting is deliberate rather than an error, and it is the reason the hydrophobic and aromatic percentages are not independent of one another. Any attempt to add the four class percentages together and check the total against one hundred will fail for any sequence containing a phenylalanine, a tryptophan or a tyrosine.

The cleanest way to see the arithmetic is a sequence of nothing but aromatic residues, where hydrophobic and aromatic are both one hundred and charged and polar are both zero. Add a single tyrosine to a long chain of aliphatic residues and all three of hydrophobic, aromatic and polar move at once, by amounts that have nothing to do with each other. The class lists overlap by design because one side chain genuinely belongs to several useful descriptions. The polar residue fraction calculator counts tyrosine as polar for exactly that reason, and the charged fraction calculator is the one whose classes do not overlap with these.

Changing a single residue moves this percentage by one hundred over n, which is 25 points on a four residue chain, 10 points on a ten residue chain and 2.5 points on a forty residue chain. On a short peptide the resolution of the count is therefore coarse, and a one residue difference between two otherwise identical peptides can move the reported figure by a quarter of the whole. Comparisons between peptides of very different lengths are correspondingly weak, and a difference of a few percent between a short chain and a long one may be nothing more than the difference in length.

This is the same arithmetic that makes a mean over a short chain unreliable, and it is why the score on the GRAVY page is best read as a sorting device rather than a measurement. Both quantities rest on the same underlying table of residue classes. The guide to the GRAVY score is useful here because it sets out the per residue values that the count and the mean both depend on.

Every residue in this class is non polar in the sense that its side chain carries no formal charge and no hydrogen bonding group, and that is a defensible line to draw. It is also only a line, and a different one would be defensible too. Glycine and proline are counted here although the behaviour of their side chains in water is more complicated than a single label suggests, and a classification that is convenient for tabulation is not automatically the most informative one for a given question.

What the count is good for is a quick, reproducible statement about the chemical character of a sequence and a comparison between sequences described on the same basis. What it is not is a prediction. A high hydrophobic fraction is associated with poor aqueous solubility and with a tendency to aggregate, and the association is real enough to be worth knowing about, but it is an association rather than a threshold, and the concentration at which anything happens is not in this number. The net charge calculator usually changes the interpretation more than this one does, because charge rather than hydrophobicity governs behaviour at moderate pH.