Hydrophobic moment

Eisenberg's hydrophobic moment along the chain, at a helical angle you choose — which separates a hydrophobic stretch that is amphipathic from one that is uniformly buried.

What it doesCopy Anchor Link

Mean hydrophobicity says how hydrophobic a stretch is. It cannot say whether the hydrophobic residues are on one face. That distinction is the whole difference between a transmembrane helix, which is hydrophobic all round, and a surface-seeking amphipathic helix, which has a hydrophobic face and a polar one — and the two behave nothing alike.

The hydrophobic moment is the vector sum of residue hydropathies placed around a helical wheel at a fixed angle per residue. A stretch whose hydrophobic residues cluster on one face gives a large moment; the same residues distributed evenly cancel and give a small one. Plotted per window along the chain, it says where the amphipathic stretches are.

The angle is a control, not a constant. 100° per residue is the α-helix, 160°–180° the β-strand, and sweeping it is how you find which periodicity a stretch actually has — a peak at 100° and nothing at 160° is a helix, and the reverse is a strand.

Each window is also classified by where the pair of values (mean hydrophobicity, moment) falls on Eisenberg's plot: globular, surface-seeking, or membrane-associated. Runs of one class are listed as segments rather than as single windows, with the minimum run length yours to set — thresholding a run into a "real" segment is a judgement, and the tool does not make it silently.

When you'd use thisCopy Anchor Link

• To tell a transmembrane helix from an amphipathic one. Both are hydrophobic on a hydropathy profile; only one has a large moment.
• To find surface-seeking and lytic segments — antimicrobial peptides, lipid-binding helices, and the amphipathic helices of many membrane-remodelling proteins are exactly this signal.
• To decide the periodicity of a stretch by sweeping the angle, rather than assuming a helix.
• To place a fusion or a tag without cutting through an amphipathic face.
• To choose which runs to select on the sequence and inspect the residues themselves.

What this computesCopy Anchor Link

• The hydrophobic moment μH per window, as the magnitude of the vector sum of residue hydropathies at the chosen angle per residue (Eisenberg's formulation).
• The mean hydrophobicity of the same window, because the classification needs both.
• The Eisenberg membrane-position class of each window — globular, surface-seeking, or membrane-associated — from where the pair falls on his plot.
• Runs of a single class, at a minimum run length you set, reported as residue ranges you can select on the sequence.
• μH at the angle currently chosen, alongside the values at the two canonical angles, so a sweep can be read against the standard cases.

What it does not doCopy Anchor Link

It assumes a regular periodic structure in the window it is scoring: a moment computed at 100° per residue is the moment the stretch would have if it were an α-helix, whether or not it forms one. It is not a structure prediction and not a topology prediction — it says a stretch has the periodicity and the asymmetry of an amphipathic helix, not that the protein contains one.