Wam logo
Home
Background
Introduction
WAM canonical classes
H3 structure
WAM algorithm
Sidechain modelling
Screens and sample results
Sequence / structure rules in screening
Submit!
WAMpredict
Download
History
Links
References
Help
CPAD home
WAM background

As has been seen on the main algorithm page , only the framework and conserved canonical sidechains are built early on. Those remaining are the canonical non-conserved sidechains, and the non-canonical sidechains, which, being highly variable from structure to structure, require special, energy-based, theoretical methods to build.

(Note, however, that for kinked H3 loops, the four C-terminal sidechains are conserved in chi-1, and chi-2 to some extent; so before building the remainder, these are built using the average chi angles in known kinked structures).

These non-conserved residues are built simultaneously. Two methods can be used; the dead-end elimination algorithm of Lasters et al ( Lasters et al, 1995; Desmet et al, 1992 ) which uses rotamers from a library (De Maeyer et al, 1997) to attempt to find the global energy minimum by rejecting rotamers which always have a higher interaction energy with the other sidechains compared to another rotamer; and the CONGEN iterative algorithm of Bruccoleri and Karplus which sequentially builds each sidechain in the lowest energy conformation with respect to the environment, then iteratively adjusts each conformation to take into account the changed environment (which will now include sidechains).
As can be seen here, the accessibility score screening method for H3 loops requires that sidechains are built first.

Advantages and disadvantages of both methods

The dead-end algorithm is slightly more sound a method in that it considers all pairs of rotamers and is less biased by an (arbitrary) initial conformation in the way CONGEN is, and uses observed rotamers rather than a more arbitrary 30 degree torsional sampling grid. However, the dead-end algorithm is three to four times slower, and tests have shown little difference in the accuracy of models. Therefore, CONGEN is a recommended option in the screens which require sidechain building on more than one final conformation, i.e. the accessibility profile screen and the VFF sidechain energy screen.

It should also be said that large numbers of aromatic residues slow the dead-end method down (as they have large numbers of rotamers) so in these cases CONGEN is particularly preferred over dead-end for the accessibility and VFF sidechain screens.

Last updated 22/11/01