DSSP

What is DSSP?

DSSP is a coordinate-based program for assigning protein secondary structure. Wolfgang Kabsch and Christian Sander introduced the method to give structural biologists a consistent, physically motivated classification based on backbone hydrogen bonding and local geometry. It analyzes an existing three-dimensional model; it does not predict a fold from sequence.

ProteinIQ runs mkdssp 4.5.8, the maintained DSSP implementation. DSSP 4 uses annotated mmCIF as its primary result format, while retaining the fixed-width legacy DSSP format for structures whose identifiers fit its limits. The mmCIF result preserves the submitted atomic coordinates and adds both standard secondary-structure records for molecular graphics and detailed dssp_ categories for quantitative analysis.

DSSP is useful when secondary-structure assignments must be reproducible across experimental models, predicted structures, molecular-dynamics snapshots, or related conformations. Because the assignments come from coordinates, differences between runs can reveal genuine local structural changes, but can also reflect missing atoms, coordinate quality, or different DSSP versions.

How to use DSSP online

Submit one protein structure as a PDB or mmCIF file, a gzip-compressed coordinate file, or an RCSB PDB ID. ProteinIQ runs DSSP 4.5.8 and returns a residue-level assignment table, structure statistics, an annotated mmCIF, an interactive structure view, CSV projections, and the optional legacy DSSP file.

Inputs

DSSP analyzes one model and all its chains together, preserving inter-chain hydrogen bonds and sheets. When you upload a PDB containing multiple models, choose Model to analyze in the structure controls, then select Apply trim. Model 1 is selected by default. The selected model's coordinates and residue numbering are preserved; you can return to the original file with Reset. mmCIF inputs and API submissions use model 1. PDB files must be valid coordinate files and include records required by mkdssp, including CRYST1. Residues without the backbone atoms needed for assignment may be absent from the residue table.

Settings

Results

The Structure view does not show a newly predicted conformation. DSSP leaves the atomic geometry intact and reannotates the standard _struct_conf, _struct_sheet, and _struct_sheet_range categories that molecular viewers use. The detailed eight-state assignment and numeric descriptors remain available in the Residues table.

How DSSP works

DSSP identifies recurring patterns in backbone hydrogen bonds, then combines those patterns into secondary-structure elements. Repeated turns form helices. Repeated bridges form ladders, and connected ladders form beta sheets. Local Cα geometry supplies complementary bend and chirality descriptors. DSSP 4 also recognizes polyproline II helices.

The one-character assignment summarizes a richer set of per-residue observations:

DSSP assignment is deterministic for the same coordinate model, version, and settings. It is still a classification of a particular model, not a statement that the protein always occupies that conformation.

Understanding the results

Residue-level descriptors

Accessibility depends on the complete coordinate context. In a multichain complex, neighboring chains can bury surface that would be exposed in an isolated monomer. DSSP ignores explicit water and other hetero atoms during this calculation, and incomplete or unusual residues can produce values that differ from expectations. For a dedicated surface-area calculation with its own output model, compare the result with SASA Calculator.

Structure statistics

The Statistics view preserves the categories and units reported by DSSP rather than collapsing them into a single score. A coordinate gap that breaks a peptide chain can increase the DSSP chain-segment count even when residues keep the same mmCIF chain identifier. Hydrogen-bond rows include raw counts and, where available, counts per 100 residues so structures of different sizes can be compared more carefully.

Comparing structures

Secondary-structure differences are most meaningful when the structures were prepared consistently and analyzed with the same DSSP version and settings. A code change near a missing residue or unresolved loop may reflect model completeness rather than a biological transition. MolProbity can help identify geometry and model-quality problems before interpretation; PDB Fixer is appropriate when missing atoms or residues must be repaired before rerunning the analysis.

For sequence-only secondary-structure estimates, use Chou-Fasman. If no coordinate model exists yet, ESMFold can generate a structure that can then be classified with DSSP. These methods answer different questions: prediction proposes structure from sequence, while DSSP assigns structural states to coordinates that already exist.