Superpositions and alignments in YASARA
In addition to a classical superposition that minimizes the RMSD between two atom selections,
YASARA supports a number of special features:
- Chemically equivalent atoms can be flipped automatically to minimize the RMSD. For example when calculating the heavy-atom RMSD between two proteins, it may happen that a certain Phe side-chain has been flipped by 180 degrees, i.e. the CD1/CE1 atoms have swapped place with CD2/CE2. Chemically the result is identical, yet a wrong RMSD will be calculated if the atoms are not flipped back. YASARA supports this flipping for all kinds of molecules by analyzing the structure to search for equivalent atoms to flip.
- Simultaneous superposition of multiple objects on their core regions. When calculating an NMR structure, it is usually needed to superpose the resulting bundle of structures all together, putting most emphasis on the core regions, so that the superposition is not negatively influenced by floppy tails. To achieve this goal, YASARA features the THESEUS maximum likelihood method.
- Structural alignments. If it is not known which atoms should actually be superposed, one needs to create an alignment first. When working with proteins, the most successful approach is to ignore the sequence and concentrate on the structure only. This allows to superpose also distantly related proteins with high accuracy. A large number structural alignment programs have been developed. YASARA has a built-in MUSTANG module and a local plugin for SHEBA in Linux and MacOSX.
- Twisted structural alignments. YASARA can bend and wind proteins to maximize the number of structurally aligned residues, for example to create sequence profiles for homology modeling.
- Structural alignments of small molecules. YASARA Model and above can automatically locate the largest common substructure of small molecules like ligands and superpose them.
- Multiple structural alignments. If the goal is to align and superpose an entire protein family, the required multiple structural alignment can be performed with the built-in MUSTANG module.
- Identify related structures in the PDB. An interface to the RCSB's weekly updated all-against-all comparison of PDB entries makes it trivial to retrieve and align all structurally related proteins to a given input structure.
- Sequence alignments. In some rare applications, one needs to superpose proteins based on a sequence alignment. YASARA supports local Smith&Waterman and global Needleman&Wunsch sequence alignments to achieve this goal.
R E F E R E N C E S
 THESEUS: maximum likelihood superpositioning and analysis of macromolecular structures
Theobald DL, Wuttke DS (2006) Bioinformatics 22 ,2171-2172
MUSTANG: A multiple structural alignment algorithm
Konagurthu AS, Whisstock JC,
Stuckey PJ, Lesk AM (2006) Proteins 64,559-574
 J.Jung & B.Lee,
(2000) Protein Eng. 13,535-543
 I.N.Shindyalov & P.E.Bourne
(1998) Protein Eng. 11,739-747
 M.Shatsky,R.Nussinov & H.J.Wolfson
(2002) Lect.Notes in Comp.Sci. 2452,235-250