CPPred Instructions


The method is based on calcuating transfer energy ΔG(z) of a peptide from water to different positions within the lipid bilayer by the PPM 2.0 method (OPM). using the implicit solvent model of the DOPC bilayer (PMID:21438606 PubMed).

The method either uses the predefined structure(s) of a peptide or predicts the formation of membrane-bound α-helix similarly to the FMAP method (FMAP).

The optimal translocation pathway of the peptide across the lipid bilayer is calculated and represented by the ΔG(z) curve. The permeability coefficient, $P_{calc}$, is determined by integrating the curve:

$\frac{1}{P_{calc}} = \int^d_0 \frac{dz}{K(z)D}$

$K(z) = e^{\frac{-\Delta{}G(z)}{RT}}$

where K(z) is depth-dependent partition coefficient from water to the bilayer; D is diffusion coefficient; z is the translational coordinate of the molecule across the lipid bilayer; and ΔG(z) is transfer energy of the molecule from water to the position z along the bilayer normal.

The calculations usually take a few (< 3) minutes. The program is running on a single slow processor. Current version of the server is only applicable to planar DOPC bilayer. Future versions will include a set of different membrane types and will analyze the ability of a peptide to induce membrane deformations, such as membrane thinning and curvature changes. They will also be applicable to β-sheet peptides.


The user has two options to upload data:

  1. Upload Amino Acid Sequence. Upload an amino acid sequence of a peptide (single-letter code) together with a peptide name (avoid spaces, periods or special symbols in the name) that will be used for the output.
  2. Upload PDB File. Upload one or several predefined 3D models of the peptide in the PDB format.

The server allows selection of the the membrane type (only DOPC bilayer in current version), and choice of experimental conditions (temperature, and pH).

The user can choose between two optimization options for calculation of the lowest transfer energy pathway ∆G(z) of the molecule across the membrane:

  1. Dragging. "Drag" the molecule across the membrane with local energy minimization with respect of rotational variables of the molecule in every point z+∆z, starting from the optimal rotational orientation in the previous point z.
  2. Global Energy Optimizaiton. Global energy optimization of rotational orientation of the molecule in each position z along the bilayer normal.

Option (a) should be used for estimating the permeability coefficient because it allows more precise estimation of the transmembrane energy barrier. Option (b) may be used for additional analysis of behavior of the peptide in membrane.

Output Information

Output includes the following data:

  1. Graphical representation of the calculated transfer energy curve, ∆G (z).
  2. Log of calculated Permeability Coefficient.
  3. Interactive 3D visual images of a given peptide along the translocation pathway (link to GLMol) and at the lowest energy arrangement in membrane (link to JMol).
  4. Two downloadable coordinate files: (a) fist file (“XXX.pdb”) represents one peptide conformation/orientation at the lowest energy arrangement in membrane; (b) second file (XXXout.pdb”) represents multiple conformations/orientations of a peptide moving across the lipid bilayer.
  5. Output messages specify calculated parameters of a membrane-bound peptide: positions in amino acid sequence of a predicted α-helix, α-helix stability in membrane, its membrane binding energy, optimal membrane penetration depth, tilt angle, and data for the ∆G(z) curve.

Interpretation of Results

  1. Values of $log(P_{calc}) > -5$ indicate that the peptide may readily penetrate across the lipid bilayer.
  2. GLMol link provides the interactive 3D images of a peptide moving across the membrane. An additional visualization can be obtained via PyMOL open source system (PyMOL) using “load XXXout.pdb, multiplex=1” command for the second output coordinate file.


Please send any questions or requests to Andrei Lomize (almz@umich.edu)