Structural Immunoinformatics - Immunoinformatics - page 54

procedure is described in an earlier study (Sali and Blundell 1993). A brief outline of
the comparative modeling process is as follows:

i. Searching for a suitable template structure from the PDB using the target
sequence as a query. The template structure is a sequence with known
structure that is significantly similar to the target sequence.
ii. Align the target sequence with the template sequence to maximize the
structural similarity using either a local-similarity dynamic programming
approach (Smith and Waterman 1981) or a global-similarity approach
(Needleman and Wunsch 1970).
iii. Substitute amino acid side chains in the template with the corresponding
ones from the target sequence.
iv. Model weakly conserved regions such as insertions/deletions and loops
between the target and template sequences.

v. Perform energy minimization to improve the stereochemistry of the

modeled structure.

Generally, modeled structures are as close to the target structure as their
templates (Sanchez and Sali 1997). This is a nontrivial achievement due to the
existence of many residue substitutions, deletions, and insertions between the target
and template sequences that must be taken into account during comparative
modeling. When several templates are selected for modeling, it is possible to
generate a model that is significantly closer to the target structure than any of the
templates as the model tends to inherit the most conserved regions from each
template (Sanchez and Sali 1997).

3.4.2 Docking Algorithm

Computer-simulated ligand binding or docking is a powerful technique for
investigating intermolecular interactions. In general, the purpose of docking
simulation is twofold: (i) to find the most probable translational, rotational, and
conformational juxtaposition of a given ligand-receptor pair and (ii) to evaluate the
relative goodness-of-fit or how well a ligand can bind to the receptor. Here, we
introduce a highly accurate docking protocol for the modeling of bound peptide
ligands to the MHC receptor. The methodology presented here is applicable to the
design of both subtype-specific vaccines as well as promiscuous peptide epitopes.

3.4.3 The Peptide Docking Procedure

Beginning with the sequence of the ligand for which the structure is to be generated
(herein referred to as the target peptide), and the availability of the target MHC
receptor structure, our docking protocol consists of three essential steps: (i) rigid
docking of residues at the ends of binding groove; (ii) loop closure of central
residues by satisfaction of spatial constraints; (iii) followed by ab initio refinements
of backbone and ligand interacting side chain. The general flow of the docking
protocol is illustrated in Fig. 2.