Method for aligning molecules in three dimensions based upon their correspondence to an exemplary template molecule for use in performing 3D QSAR analyses
View Patent ↗A computerized procedure for aligning molecules for use in CoMFA or other 3D QSAR methodologies does not rely on fragmentation of the molecules instead, aligning molecules based upon comparison to identified template molecules. Initially an anchor bond is identified in the candidate and template molecule that have similar atoms at each end of the bonds. The anchor bond need not be an acyclic bond. The candidate molecule is overlaid onto the template molecule by aligning the anchor bonds. Starting and working away from the anchor bond, matching atoms or atom types between the candidate and template molecule are identified. Once all matching atoms have been identified, the 3D coordinates of the template atoms are assigned to the corresponding atoms in the candidate molecule to place the candidate molecule into alignment.
1. A computer implemented drug discovery method utilizing a specifically programmed computer for aligning in three dimensions a candidate molecule exhibiting a common activity to a specified template molecule comprising the following steps:
specifying a candidate molecule that is to be aligned to a template molecule;
generating a 3D structure for the candidate molecule determining valances, bond angles, and bond lengths;
specifying a template molecule with an identified 3D structure;
specifying an anchor bond in the candidate molecule;
specifying an anchor bond in the template molecule;
orienting the candidate molecule by superimposing its anchor bond onto the template anchor bond;
enumerating for both template and candidate molecule all atoms and atom properties excluding all hydrogen atoms;
matching similar atoms or atom properties in the template and candidate molecule by sequential traversals of the molecules that start at the anchor bond and ending wherever no more property matches exist along any connected path;
assigning the coordinates of template atoms to the matching candidate atoms;
positioning any atoms in the candidate molecule that do not have matching atoms in the template molecule by assigning coordinates to those atoms so as to maintain their relative positions to those atoms in the candidate molecule that had matching atoms in the template molecule;
aligning and assigning coordinates to atoms in any branch that includes only uncoordinated-copied atoms using the topomer alignment rules or any similar systematic procedure to consistently place the atoms in the branch; and
outputting the coordinates for the candidate molecule alignment.
2. The method of claim 1 , in which a CoMFA analysis is performed based upon a data table consisting of molecular shape descriptors generated using the resulting molecular alignment along with the activity data associated with each aligned molecule.
3. A computer implemented drug discovery method utilizing a specifically programmed computer for aligning in three dimensions candidate molecules exhibiting a common activity to a specified set of template molecules comprising the following steps:
a) specifying a candidate molecule that is to be aligned to a template molecule;
b) generating a 3D structure for the candidate molecule determining valances, bond angles, and bond lengths;
c) specifying a template molecule with an identified 3D structure;
d) specifying an anchor bond in the candidate molecule;
e) specifying an anchor bond in the template molecule;
f) orienting the candidate molecule by superimposing its anchor bond onto the template anchor bond;
g) enumerating for both template and candidate molecule all atoms and atom properties excluding all hydrogen atoms;
h) matching similar atoms or atom properties in the template and candidate molecule by sequential traversals of the molecules that start at the anchor bond and ending wherever no more property matches exist along any connected path;
i) assigning the coordinates of template atoms to the matching candidate atoms identified in step h;
j) positioning any atoms in the candidate molecule that do not have matching atoms in the template molecule by assigning coordinates to those atoms so as to maintain their relative positions to those atoms in the candidate molecule that had matching atoms in the template molecule;
k) aligning and assigning coordinates to atoms in any branch that includes only uncoordinated-copied atoms using the topomer alignment rules or any similar systematic procedure to consistently place the atoms in the branch;
l) retaining the alignment if it is better than any proceeding alignment;
m) repeating steps e) through I) for all bonds in the template molecule;
n) repeating steps d) through m) for all bonds in the candidate molecule;
o) repeating steps c) through n) for all template molecules;
p) repeating steps a) through o) for all candidate molecules; and
q) outputting the coordinates for each candidate alignment.
4. The method of claim 3 in which a CoMFA analysis is performed based upon a data table consisting of molecular shape descriptors generated using the resulting molecular alignments along with the activity data associated with each aligned molecule.