- Document Number:
20220383992
- Appl. No:
17/606960
- Application Filed:
July 17, 2019
- نبذة مختصرة :
There is provided a method for a machine learning based method of analysing drug-like molecules by representing the molecular quantum states of each drug-like molecule as a quantum graph, and then feeding that quantum graph as an input to a machine learning system.
- Claim:
1-17. (canceled)
- Claim:
18. A machine learning based method of modelling a thermodynamic ensemble or representation of a drug-like molecule, in which a sample of a thermodynamic ensemble or representation is synthetically generated and inputted into a machine learning system, the thermodynamic ensemble or representation being a molecular orbital representation or quantum graph representation of the drug-like molecule.
- Claim:
19. Method of claim 18, in which every element of the thermodynamic ensemble can be represented as a Q-graph or tensor network, or molecular orbital representation.
- Claim:
20. Method of claim 18, in which a quantum graph is a molecular graph representation of a molecule obtained from quantum mechanical calculations.
- Claim:
21. Method of claim 18, in which a quantum graph is a molecular graph representation in which each node corresponds to a molecular orbital and the edges correspond to a type of quantum correlation between these orbitals.
- Claim:
22. Method of claim 18, in which a quantum graph depends on the conformational state of a molecule.
- Claim:
23. Method of claim 18, in which a molecular orbital representation is a tensor network representation of molecular quantum states of a drug like molecule.
- Claim:
24. Method of claim 18, in which the synthetic generation of one or more samples of thermodynamic ensembles or representations is based on a thermodynamic quantity.
- Claim:
25. Method of claim 18, in which the machine learning system is configured to output a thermodynamic quantity based on an approximate expectation value over the entire or a representative set of the modelled thermodynamic ensemble or representation.
- Claim:
26. Method of claim 18, in which the machine learning system is configured to learn the distribution of Boltzmann weights of the entire or a representative set of the thermodynamic ensemble or representation of the molecule.
- Claim:
27. Method of claim 18, in which determining the cost function or backpropagation of the machine learning system is based on a thermodynamic quantity to be outputted by the system.
- Claim:
28. Method of claim 18, in which the size of the synthetically generated sample of the thermodynamic ensemble or representation is tuned depending on a downstream application.
- Claim:
29. Method of claim 18, in which the machine learning system is a graph convolutional neural network.
- Claim:
30. Method of claim 18, in which the generated sample of thermodynamic ensemble or representation is inputted as a molecular graph such as a quantum graph.
- Claim:
31. Method of claim 18, in which the machine learning system is configured to output any quantity that is a function of a thermodynamic ensemble or representation, such as binding affinity, lipophilicity, thermodynamic solubility, kinetic solubility, melting point or protonation.
- Claim:
32. Method of claim 18, in which the machine learning system is used to predict ligand protein binding affinity, and in which synthetically generated samples of the thermodynamic ensemble or representation of the ligand in solution, of the protein in solution, and of the ligand-protein complex are inputted into the machine learning system.
- Claim:
33. Method of claim 18, in which the machine learning system is used to predict ligand protein inhibition concentration, and in which synthetically generated samples of the thermodynamic ensemble or representation of the ligand in solution and of the ligand-protein complex are inputted into the machine learning system.
- Claim:
34. Method of claim 18, in which the machine learning system is used to predict lipophilicity of a drug-like molecule, and in which synthetically generated samples of the thermodynamic ensemble or representation of the unionized and/or ionized state of the molecule in octanol, and in water are inputted into the machine learning system.
- Claim:
35. Method of claim 18, in which the machine learning system is used to predict thermodynamic solubility of the drug-like molecule, and in which synthetically generated samples of the thermodynamic ensemble or representation of the solid state of the molecule and of the dissolved state of the molecule are inputted into the machine learning.
- Claim:
36. Method of claim 18, in which the machine learning system is used to predict kinetic solubility of the drug-like molecule, and in which synthetically generated samples of the thermodynamic ensemble or representation of the amorphous solid state of the molecule and of the dissolved state of the molecule are inputted into the machine learning.
- Claim:
37. Method of claim 18, in which the machine learning system is used to predict melting point of the drug-like molecule, and in which synthetically generated samples of the thermodynamic ensemble or representation of the solid state of the molecule and of the molecule in octanol are inputted into the machine learning.
- Claim:
38. Method of claim 18, in which the machine learning system is used to predict pKa of the drug-like molecule and in which synthetically generated samples of the thermodynamic ensemble or representation of the small drug-like molecule in the appropriate environment, such as water or pocket on a target protein, are inputted into the machine learning.
- Claim:
39. Method of claim 18, in which the machine learning system uses GAN or VAE, or GCPN style models.
- Claim:
40. Method of claim 18, in which the machine learning system uses generative models to learn new thermodynamic ensemble or thermodynamic ensemble for which data is not available.
- Claim:
41. Method of claim 18, in which the machine learning system implements weight sharing method when multiple generated samples of thermodynamic ensembles or representations are inputted into the machine learning system.
- Claim:
42. Method of claim 18, in which docking is used to generate the sample of thermodynamic ensemble or representation.
- Claim:
43. Method of claim 18, in which docking enhanced by molecular dynamics is used to generate the sample of thermodynamic ensemble or representation.
- Claim:
44. A machine learning based system configured to model a thermodynamic ensemble or representation of a drug-like molecule, in which the system is configured to receive and process a synthetically generated sample of a thermodynamic ensemble or representation, the thermodynamic ensemble or representation being a molecular orbital representation or quantum graph representation of the drug-like molecule.
- Claim:
45-64. (canceled)
- Claim:
65. A molecule or class of drug-like molecules identified using a machine learning based method of modelling a thermodynamic ensemble or representation of a drug-like molecule, in which a sample of a thermodynamic ensemble or representation is synthetically generated and inputted into a machine learning system, the thermodynamic ensemble or representation being a molecular orbital representation or quantum graph representation of the drug-like molecule.
- Current International Class:
16; 06; 16; 06
- الرقم المعرف:
edspap.20220383992
Processing Request
No Comments.