Secondary structure analysis using DSSP
DSSP (Dictionary of Secondary Structure of Proteins) is one of the most widely used algorithms for assigning secondary structures to protein residues based on a 3D structure (PDB coordinates)
The DSSP algorithm was originally described in the following paper:
Kabsch W, Sander C (1983). “Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features”. Biopolymers 22 (12): 2577-637. doi:10.1002/bip.360221211
Please also check another algorithm STRIDE: [[STRIDE for secondary structure]].
1. Input
- The algorithm takes atomic coordinates of a protein (N, Cα, C, O atoms) from a PDB file or trajectory frame.
- DSSP primarily focuses on backbone atoms to determine the local geometry.
2. Hydrogen bond detection
Secondary structure is largely determined by hydrogen bonding patterns in the backbone:
- DSSP uses an electrostatic model to estimate H-bond energies.
- The energy between a donor N-H and acceptor C=O is calculated using:
where $r_{XY}$ are distances between atoms.
- If the H-bond energy is below a threshold (≈ -0.5 kcal/mol), a hydrogen bond is considered present.
3. Helix and sheet assignment
DSSP assigns secondary structures based on hydrogen bonding patterns and backbone geometry:
| Structure | Definition |
|---|---|
| H (α-helix) | Residue i forms H-bonds with i+4 (C=O…H-N) |
| G (3-10 helix) | Residue i forms H-bonds with i+3 |
| I (π-helix) | Residue i forms H-bonds with i+5 |
| E (extended β-strand) | Part of β-sheet; participates in inter-strand H-bonds |
| B (isolated β-bridge) | Single H-bonded residue not in extended sheet |
| T (turn) | Residue forms H-bond with i+3 (like 3-10) but not regular helix |
| S (bend) | Geometric bend without H-bonds |
| C / ‘-’ | Coil / other / unassigned |
- DSSP looks for patterns of hydrogen bonds along the backbone.
- For example, a stretch of 4 residues with i→i+4 H-bonds is assigned as α-helix (H).
- β-strands are assigned if residues participate in inter-strand hydrogen bonds, forming a β-sheet.
4. In practice
DSSP has been implemented into the MDAnalysis package:
https://docs.mdanalysis.org/dev/documentation_pages/analysis/dssp.html
[!NOTE] Caveat Since DSSP considers geometry of the backbone of residues for assigning secondary structures, for example,
i-i+4for hydrogen bond detection, we need to select upstream and downstream at least +5 residues for calculating secondary structures.
For example:
if we want to calculate the secondary structures of the segment of residue index 10 to 20, we need to at least select residue index 5 to 25.
With the above being said, the starting and the ending residues of your selected segment will almost always be assigned as “C” (coil).