METHOD OF SEARCHING NOVEL LIGAND COMPOUNDS FROM THREE-DIMENSIONAL STRUCTURE DATABASE

Patent №

US 6,389,378

Granted

2002-05-14

Filed 1997

Owner

Lab

AI components

1

evo

Assignment

None on record

Dataset

AIPD

2023_r1 edition

Application

08817646

A method of searching one or more ligand-candidate compounds to a target biopolymer from a three-dimensional structure database, which comprises the steps of: (i) the first step of assigning hydrogen-bonding category numbers, information for calculating force-field energy, and information for generating conformations to two or more trial compounds in addition to atomic three-dimensional coordinates thereof; (ii) the second step of preparing physicochemical information about a ligand-binding region and one or more dummy atoms based on the three-dimensional atomic coordinates of the target biopolymer;(iii) the third step of estimating the most stable docking structure, wherein said step further comprises the steps of examining possible docking structures by docking a trial compound to the biopolymer while varying conformations of the trial compound, evaluating interaction energies between the target biopolymer and the trial compound, and repeating structure optimization; (iv) the fourth step of deciding whether or not the trial compound should be adopted as a ligand-candidate compound based on given criteria including the interaction energy values between the target biopolymer and the trial compound in the most stable docking structure estimated according to the step (iii); and (v) the fifth step of repeating the step (iii) and the step (iv) for all of the trial compounds.

Evolutionary computationG16C 20/40G16B 15/30G16B 15/00Y10S 707/99936Y10S 707/99945

AI classification

Evolutionary computation0.89
Planning0.15
Machine learning0.06
Natural language0.04
AI hardware0.02
Vision0.00
Knowledge representation0.00
Speech0.00
© 2026 NYSGPT2525 LLC