# US 2025/0095793 A1 · Methods and systems for studying molecules and properties thereof

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0095793 A1
- Document ID: US20250095793A1
- Status: Published application
- Kind: A1
- Application number: 18/814,910
- Filed: 2024-08-26
- Published: 2025-03-20
- Earliest claimed: 2022-02-28
- Assignees: Good Chemistry Inc.
- Applicants: Good Chemistry Inc.
- Inventors: Takeshi YAMAZAKI; Kevin RYCZKO; Arman ZARIBAFIYAN; Nima ALIDOUST
- Web page: https://patentgrep.com/patent/US20250095793A1

## CPC

- G16C 20/30: Prediction of properties of chemical compounds, compositions or mixtures
- G16B 15/20: Protein or domain folding
- G16B 15/30: Drug targeting using structural data; Docking or binding prediction
- G16B 20/30: Detection of binding sites or motifs
- G16B 40/20: Supervised data analysis
- G16C 20/50: Molecular design, e.g. of drugs
- G16C 20/70: Machine learning, data mining or chemometrics
- G16C: COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16B: BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY

## Abstract

Provided herein is a method for studying molecule and properties thereof, the method comprising: a) obtaining a request comprising an indication of at least one property of a molecule and a corresponding task; b) performing inference on at least one machine learning (ML) model using said indication, the at least one machine learning (ML) model configured to mimic result of said task to generate an inference outcome; c) performing inference reliability test on the inference outcome; i) obtaining task result using the inference outcome in response to a satisfactory inference reliability test; ii) performing said task using said indication to obtain task result in response to a nonsatisfactory inference reliability test; d) outputting said task result.

## Claims

### Claim 1

A method for studying molecule and properties thereof, the method comprising:
a) obtaining a request comprising an indication of at least one property of a molecule and a corresponding task;
b) performing inference on at least one machine learning (ML) model using said indication, the at least one machine learning (ML) model configured to mimic result of said task to generate an inference outcome;
c) performing inference reliability test on the inference outcome;
i) obtaining task result using the inference outcome in response to a satisfactory inference reliability test;
ii) performing said task using said indication to obtain task result in response to a nonsatisfactory inference reliability test;
d) outputting said task result.

### Claim 2 (depends on 1)

The method of claim 1, further comprising training the at least one ML model using a training dataset, wherein said training dataset comprises said task result obtained in c) ii) in response to the nonsatisfactory inference reliability test.

### Claim 3 (depends on 1)

The method of claim 1, further comprising, storing, in response to the nonsatisfactory inference reliability test, in a database said indication of said at least one property of said molecule, said corresponding task, and said task result obtained in c) ii).

### Claim 4 (depends on 3)

The method of claim 3, further comprising, prior to b), outputting a task result if said indication of said at least one property of said molecule, said corresponding task and said task results are stored in said database.

### Claim 5 (depends on 1)

The method of claim 1, further comprising, prior to b), obtaining at least one of accuracy requirements and limit for computational resources.

### Claim 6 (depends on 1)

The method of claim 1, wherein said indication of at least one property of a molecule comprises at least one member of the group consisting of three dimensional coordinates of the atoms on the molecule, a fingerprint, simplified molecular-input line-entry system (SMILES), International Chemical Identifier (InChI), description of the shape of the spectroscopic, and name of the property.

### Claim 7 (depends on 1)

The method of claim 1, wherein said at least one property comprises at least one member of the group consisting of molecular coordinate, ground state energy, excited states energies, highest occupied molecular orbital (HOMO)-lowest unoccupied molecular orbital (LUMO) gap, lonization potential, electron affinity, singlet-triplet gap, atomic charge, dipole moment, charge density, spectroscopic properties, peak position at X nm wherein X is the peak position, and binding affinity with a target molecule, equilibrium geometry, reactivity, hydrophobicity, synthesizability, conformational entropy, and residence time of a molecule interacting with another molecule.

### Claim 8 (depends on 1)

The method of claim 1, wherein said task comprises at least one member of the group consisting of protein structure prediction, protein pocket identification, virtual screening based on protein-ligand interactions, quantitative structure-activity relationship prediction, molecular similarity search, classical force field generation, classical molecular dynamics simulations, binding free energy calculations, toxicity prediction, synthesizability prediction, generation of candidate catalyst molecules, conformational search, virtual screening based on the outcome prediction and the barrier height, prediction for the reaction between catalyst candidate and substrate, ab initio molecular dynamics simulations, reaction barrier height calculation, molecular coordinate calculation, ground state energy calculation, excited states energies, HOMO-LUMO gap calculation, ionization potential calculation, electron affinity calculation, singlet-triplet gap calculation, atomic charge calculation, dipole moment calculation, charge density calculation, spectroscopic properties prediction, equilibrium geometry prediction, reactivity prediction, hydrophobicity prediction, conformational entropy prediction, residence time of a molecule interacting with another molecule prediction.

### Claim 9 (depends on 1)

The method of claim 1, wherein said task comprises a pipeline comprising a plurality of subtasks.

### Claim 10 (depends on 1)

The method of claim 1, wherein said task result comprises at least one member of the group consisting of the number indicating the ground state energy, three-dimensional coordinates of the atoms in the molecule, and the numbers indicating the peak positions of a spectroscopy information.

### Claim 11 (depends on 1)

The method of claim 1, wherein said performing said task in c) ii) comprises at least one member of the group consisting of performing a corresponding experiment, implementing computational chemistry method, quantum chemistry method, molecular mechanics method, computing energy, computing electronic structure, optimizing molecular geometry, performing the transition state search, performing conformational search, performing molecular similarity search, performing classical molecular dynamics simulation, performing ab initio molecular dynamics simulation, performing protein structure prediction, performing protein binding site prediction, performing virtual screening, performing protein-ligand binding structure prediction, performing free energy perturbation, performing ligand optimization, performing catalyst optimization, performing reaction path prediction, performing synthesizability prediction, performing spectroscopic information prediction, performing reactivity prediction, performing toxicity prediction, performing the binding structure prediction between enzyme and substrate, performing the structure prediction of self-assembled nanomaterials, optimizing the composition of the material, optimizing the experimental condition.

### Claim 12 (depends on 11)

The method of claim 11, wherein said performing structure prediction comprises using a quantum chemistry method using at least one member of the group consisting of Hartree-Fock (HF) method, Density Functional Theory (DFT), Coupled-Cluster Single-, Double-, and perturbative Triple-excitations (CCSD(T)), Full Configuration Interaction (FCI), Heat-Bath Configuration Interaction (HBCI), Quantum Monte Carlo Full Configuration Interaction (QMCFCI), Density Matrix Embedding Theory (DMET), Fragment Molecular Orbital method (FMO), Incremental Full Configuration Interaction (iFCI), ML-based Schrodinger equation solver such as Paulinet, Hybrid quantum mechanics-molecular mechanics (QM/MM), and ab initio molecular dynamics (AIMD) simulation.

### Claim 13 (depends on 1)

The method of claim 1, wherein said at least one machine learning (ML) model comprises supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, active learning, delta learning, continuous learning, and transfer learning.

### Claim 14 (depends on 1)

The method of claim 1, wherein said inference reliability test comprises standard deviation computation using the inference outcome.

### Claim 15 (depends on 1)

The method of claim 1, wherein said inference reliability test comprises standard error computation using the inference outcome.

### Claim 16 (depends on 1)

The method of claim 1, wherein said inference reliability test comprises a similarity search between the training dataset and the request to identify a rareness of the request.

### Claim 17 (depends on 1)

The method of claim 1, wherein said inference reliability test comprises instance selection strategies for selecting critical instances to refine the ML models, wherein the instance selection strategies comprise an uncertainty sampling approach, a query by committee approach, an expected model change approach, an expected error reduction approach, and density weighted methods.

### Claim 18

A system for studying molecule and properties thereof, the system comprising:
a) a memory component, comprising instructions for studying molecule and properties thereof and at least one trained machine learning (ML) model;
b) a computational platform operatively coupled to said memory component, wherein the computational platform is configured to execute said instructions to at least:
i) obtain a request comprising an indication of at least one property of a molecule and a corresponding task;
ii) perform inference on at least one machine learning (ML) model using said indication, the at least one machine learning (ML) model configured to mimic result of said task to generate an inference outcome;
iii) perform inference reliability test on the inference outcome; obtain task result using the inference outcomes in response to a satisfactory inference reliability test; perform said task using said indication to obtain task result in response to a nonsatisfactory inference reliability test.

### Claim 19 (depends on 18)

The system of claim 18, further comprising a database configured to store tasks and results thereof, wherein said instructions comprise storing in said database said indication of said at least one property of said molecule, said task and said task result.

### Claim 20 (depends on 18)

The system of claim 18, further comprising a distributed computing system.

### Claim 21 (depends on 18)

The system of claim 18, further comprising a cloud computing system.

### Claim 22 (depends on 18)

The system of claim 18, further comprising a classical computer.

### Claim 23 (depends on 18)

The system of claim 18, further comprising a non-classical computer.

### Claim 24 (depends on 23)

The system of claim 23, wherein said non-classical computer is a quantum computer.

### Claim 25 (depends on 24)

The system of claim 24, wherein said quantum computer comprises at least one member of the group consisting of quantum devices, such as superconducting quantum computers, trapped ion quantum computers, optical quantum computers, or quantum annealer.
