Loading patent…
AssigneeArris Pharmaceutical Corporation
InventorsDavid Chapman, Roger Critchlow, Ajay N. Jain, Rick Lathrop, Tomas L. Perez, Tom Dietterich
Estimating the term…
Explicit representation of molecular shape of molecules is combined with neural network learning methods to provide models with high predictive ability that generalize to different chemical classes where structurally diverse molecules exhibiting similar surface characteristics are treated as similar. A new machine-learning methodology that can accept multiple representations of objects and construct models that predict characteristics of those objects. An extension of this methodology can be applied in cases where the representations of the objects are determined by a set of adjustable parameters. An iterative process applies intermediate models to generate new representations of the objects by adjusting said parameters and repeatedly retrains the models to obtain better predictive models. This method can be applied to molecules because each molecule can have many orientations and conformations (representations) that are determined by a set of translation, rotation and torsion angle parameters.
A method in a computer for predicting activity of molecules with respect to a chemical function based on known activities of a plurality of molecules, said molecules each having one or more conformations and orientations, each combination of a conformation and an orientation defining a pose of a molecule, said method comprising:
A method of operating a computer for predicting activity of molecules with respect to a chemical function based on known activities of a training set of molecules, said molecules in the set each having one or more conformations and orientations, each combination of a conformation and an orientation defining a pose of a molecule, said method comprising:
A method in a computer for arriving at a model for predicting activity of molecules with respect to a chemical function based on known activities of a training set of molecules, said molecules in the set each having one or more conformations and orientations, each combination of a conformation and an orientation defining a pose of a molecule, said method comprising:
A method of operating a computer for predicting characteristics of an object based on known characteristics of a plurality of other objects, said other objects each having one or more representations, said method comprising:
A method of operating a computer for predicting characteristics of an object based on known characteristics of a plurality of other objects, said other objects each having many representations, each representation called a pose and being defined by the object and values of one or more parameters defining pose parameters, said method comprising:
A method for predicting activity of molecules with respect to a chemical function based on activities of a plurality of molecules, said method comprising
Dependencies are inferred from the claim text.