General strategy for vaccine and drug design

A method for modeling the minimum energy conformation of a ligand within the binding site of a receptor molecule, comprising the steps of:

• establishing X, Y, and Z coordinate axes for the receptor molecule to be modeled so that a rectangular region is established with sides parallel to the coordinate axes with the origin of the axes lying in the binding site of the receptor molecule;
• establishing a grid in the X, Y, and Z coordinate axes with a grid mesh size of sufficient resolution to distinguish low-energy amino acid locations in the binding site;
• retrieving previously established X, Y, and Z coordinate axes for the receptor molecule and grid in the X, Y, and Z coordinate axes;
• retrieving previously established low-energy amino acid configurations within the binding site;
• selecting a ligand anchor pair from among the low-energy amino acid configurations consistent with the amino acid sequence of the ligand;
• completing the ligand structure by closing a loop between the ligand anchor pair with amino acid residues using a loop closure process;
• and
• determining a minimum energy configuration as defining the conformations of the ligand within the receptor binding site using a molecular modeling process allowing both the ligand atoms and the receptor atoms to adjust.

A method for modeling the physical properties of a binding site in a receptor molecule and determining the locations and conformations of end-residues of a ligand that binds to the receptor comprising the steps of:

• establishing X, Y, and Z coordinate axes for the receptor molecule to be modeled so that a rectangular region is established with sides parallel to the coordinate axes with the origin of the axes lying in the binding site of the receptor molecule;
• establishing a box, A, surrounding the probable, location of the amino terminus of the ligand and a box, C, surrounding the probable location of the carboxy terminus of the ligand;
• placing at random positions in box A copies of a model of the amino terminus ligand residue and placing in box C copies of a model of the carboxy terminus of the ligand;
• minimizing energy by a multi-copy mean-field approximation of the collection of amino termini located in box A by simultaneously moving all the charged end amino acid residue copies within box A until each copy obtains a minimum energy location;
• grouping the copies of the amino terminus into clusters of similar location and configuration;
• minimizing energy by a multi-copy mean-field approximation of the collection of carboxy termini located in box C by simultaneously moving all the charged end carboxy residue copies within box C until each copy obtains a minimum energy location;
• grouping copies of the carboxy terminus into clusters of similar location and configuration;
• selecting a representative of the amino terminus from the largest cluster in box A as defining the location and configuration of the amino terminus of the ligand;
• selecting a copy of the carboxy terminus from the largest cluster in box C as defining the location and configuration of the carboxy terminus of the ligand;
• and
• storing the selected locations and configurations for the amino terminus and carboxy terminus.

A method for modeling the minimum energy conformation of a ligand within the binding site of a receptor molecule, comprising the steps of:

• establishing X, Y, and Z coordinate axes for the receptor molecule to be modeled so that a rectangular region is established with sides parallel to the coordinate axes with the origin of the axes lying in the binding site of the receptor molecule;
• retrieving previously established low-energy end-residues for the amino terminus and carboxy terminus within the binding site;
• completing the ligand chain structure by closing a loop between the end-residues with models of amino acid residues corresponding to the ligand using torsion angles corresponding to the expected conformation of the ligand and scaling bond lengths to make the intervening chain of correct length to reach between the two end-residues;
• randomly generating a set of similar loops with torsion and bond angles randomly perturbed from that of the previously constructed chain;
• minimizing energy by a multi-copy mean-field approximation of the collection of model ligands by simultaneously moving all of the ligands and rescaling the bond lengths to reach known covalent bond lengths;
• grouping the collection of ligands into clusters of similar location and configuration;
• and
• selecting from the largest cluster a representative defining the predicted ligand structure.