Presented below is a short description of each method applied in 2Struc. Please click the purple buttons to access the references. In each reference box you will find a link to the original published work and, where available, web based resources. If you use 2Struc and publish your work please cite our paper (Klose, D & R.W. Janes, 2010, 2Struc - The Protein Secondary Structure Analysis Server, Biophysical Journal, 98:454a-455) and each of the methods you run.
DSSPDictionary of Secondary Structure of Proteins (DSSP) assigns eight state secondary structure using hydrogen bonds alone. DSSP defines a hydrogen bond where the bond energy is below -0.5 kcal/mol based on a Coulomb approximation of the hydrogen bond energy.
DSSPcontDSSPcont is a variant of DSSP aiming to assign secondary structure in a continuous fashion, thus reflecting the structural variability due to thermal motion. The continuous output is achieved by running DSSP multiple times with different hydrogen bond thresholds and then imposing a weighted average over the individual DSSP assignments to each residue.
StrideThe STRuctural IDEntification method (STRIDE) uses empirically derived hydrogen bond energy and phi-psi torrsion angles to assign secondary structure. Torsion angles are given propensities according to how close they are to their regions in the Ramachandran plot. STRIDE is optimised to meet the visual assignments made by crystallographers.
P-SEAP-SEA assigns secondary structure using Cα coordinates. This is achieved using a short range Cα distance mask and two angle criteria for each secondary structure.
PALSSEPredictive Assignment of Linear Secondary Structure Elements (PALSSE) describes three state secondary structure in vector form based on Cα coordinates. The helix class includes 310, pi and alpha helices as well as turns that show a helical propensity based on the observation that some helices start and end with tighter, looser and non-backbone hydrogen bonded turns and helices. The β class includes parallel/anti-parallel sheets, bridges, bends and hairpins. The net result is over-assignment of secondary structure compared to other methods with approximately 80% of residues being assigned a regular secondary structure class. While over representation could be regarded as a handicap, it proves to be useful in structure comparison and similarity searches.
STICK: Continuous assignment based on line segmentsSTICK finds a set of line segments independent of external secondary structure definition, allowing the segments to be used as a novel basis for secondary structure definition. This is achieved by taking the average rise/residue along each axis to characterise the segment. This approach has the advantage that the secondary structure elements are described by a continous value and are therefore not restricted to the conventional structure classes. This behavior allows structures absent of “classic' secondary structure to be encoded as line segments that can be applied in structure comparison algorithms. The raw output of STICK is not yet available to download.
KAKSIKAKSI assigns secondary structure based on Cα distance measures and phi-psi angles and was designed to be sensitive to to kinks in helices.
TM-AlignA tool for aligning 3D protein structures