# US 2026/0290520 A1 · Retrosynthesis prediction system and method using graph generative models

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

- Number: US 2026/0290520 A1
- Document ID: US20260290520A1
- Status: Published application
- Kind: A1
- Application number: 18/871,585
- Filed: 2023-04-05
- Published: 2026-09-24
- Earliest claimed: 2022-06-05
- Assignees: Ohio State Innovation Foundation
- Applicants: Ohio State Innovation Foundation
- Inventors: Xia NING; Ziqi CHEN
- Web page: https://patentgrep.com/patent/US20260290520A1

## CPC

- G16C 20/50: Molecular design, e.g. of drugs
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06N 3/045: Combinations of networks
- G06N 3/09: Supervised learning
- G06N 5/02: Knowledge representation; Symbolic representation
- G16B 15/30: Drug targeting using structural data; Docking or binding prediction
- G16B 40/20: Supervised data analysis
- G16C 20/10: Analysis or design of chemical reactions, syntheses or processes
- G16C 20/70: Machine learning, data mining or chemometrics
- G16C 20/80: Data visualisation
- G16C: COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16B: BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY

## Abstract

An exemplary retrosynthesis prediction method and system, e.g., for a semi-template-based approach, are disclosed that imitates the reversed logic of synthetic reactions: first predicting the reaction centers in target molecules, identifies the synthons needed to assemble the final products, and the transforms these synthons into reactants. Among other things, the exemplary retrosynthesis prediction method and system can learn from molecular graphs that encode information most directly relevant to synthetic transformations—considering not only the structures of the synthons to be completed but also the structures of other synthons from the same products so as to utilize their relations during reactions.

## Claims

### Claim 1

A method of performing retrosynthesis to predict output reactants molecules that can be used in a synthetic molecule model for an input target molecule, the method comprising:
receiving, by a processor, an input molecule data structure represented as a graph data object;
evaluating, by the processor, one or more graph data objects of the input molecule data structure, wherein evaluating includes a set of sequences to:
determine, by the processor, via one or more trained AI algorithm via a training data set of molecule data structures, for one or more reaction center types, a likelihood value that a given bond of a plurality of assessed bonds or a given atom of a plurality of assessed atoms of each of one or more graph data objects is likely a reaction center for a subsequent synthesis, wherein the one or more reaction center types include:
i) a first reaction center type associated with a new chemical bond formation that is formed during a hypothetical reaction and exists in the input target molecule but not in the output reactants molecules;
ii) a second reaction center type associated with an existing chemical bond in the input molecule data structure being changed in type during a hypothetical reaction; and/or
iii) a third reaction center type associated with an atom in the input molecule data structure connected with fragments being removed during a hypothetical reaction without a new chemical bond being formed or a type of the existing chemical bond being changed,
wherein the likelihood values of the plurality of assessed bonds of each of one or more graph data objects are used in a selection of one or more reaction centers for the one or more graph data objects to generate a plurality of intermediate molecular graph data objects each having a modified chemical bond at the selected one or more reaction centers for the one or more graph data objects, wherein one or more of the plurality of intermediate molecular graph data objects with the modified chemical bond are attached to a chemical fragment at the modified chemical bond to generate a completed reactant graph, wherein the completed reactant graph is used to generate the synthetic molecule model.

### Claim 2 (depends on 1)

The method of claim 1, wherein the one or more of the plurality of intermediate molecular graph data objects with the modified chemical bond are generated by:
determining, by the processor, a likelihood of a neighbor bond type change for a given reaction center.

### Claim 3 (depends on 2)

The method of claim 2, wherein the likelihood of the neighbor bond type change for the given reaction center is determined using a second trained AI algorithm that learns for a given bond structure and on a structure of the graph data objects of the training data set of molecule data structures.

### Claim 4 (depends on 1)

The method of claim 1, wherein the one or more of the plurality of intermediate molecular graph data objects with the modified chemical bond are generated by:
determining, by the processor, a likelihood of an atom charge change for all the atoms involved in a given reaction center and their associated neighboring bonds.

### Claim 5 (depends on 4)

The method of claim 4, wherein the likelihood of the atom charge change is determined using a third trained AI algorithm that learns, via training data set of molecule data structures, for a given atom structure probabilities of accepting one electron, donating one electron, or no electron change during a hypothetical reaction

### Claim 6 (depends on 1)

The method of claim 1, wherein the one or more of the plurality of intermediate molecular graph data objects with the modified chemical bond are generated by:
iteratively attaching, by the processor, a chemical substructure to a given intermediate molecular graph data object of the plurality of intermediate molecular graph data objects based on a fourth trained AI algorithm that learns for a given atom structure and on a structure of the graph data objects of the training data set of molecule data structures.

### Claim 7 (depends on 6)

The method of claim 6, wherein iteratively attaching the chemical substructure includes:
determining, by the processor, likelihood values for inclusion of one or more types of new substructure from a library of candidate substructures each comprising a bond or a ring structure, wherein the likelihood values are determined using a fifth trained AI algorithm that learns for a given atom structure and on a structure of the graph data objects of the training data set of molecule data structures.

### Claim 8 (depends on 1)

The method of claim 1, wherein the one or more graph data objects include a molecular graph and a BRICS graph.

### Claim 9 (depends on 1)

The method of claim 1, wherein the one or more graph data objects include a molecular graph and a junction tree.

### Claim 10 (depends on 1)

The method of claim 1, wherein the selection of one or more reactions for the one or more graph data objects includes:
generating, by the processor, a ranked list of candidate reactant graphs; and
selecting, by the processor, via a beam search, a pre-defined number of top reaction centers as the one or more reaction centers for the one or more graph data objects to generate the plurality of intermediate molecular graph data objects.

### Claim 11 (depends on 1)

The method of claim 1, wherein the likelihood value that the given bond of the plurality of assessed bonds of each of one or more graph data objects is likely the first reaction center type is determined based on a sixth trained Al algorithm, wherein the sixth trained AI algorithm operation is configured to learn, using the training data set of molecule data structures, based on a given bond structure and on a structure of the graph data objects.

### Claim 12 (depends on 1)

The method of claim 1, wherein the likelihood value that the given bond of the plurality of assessed bonds of each of one or more graph data objects is likely the second reaction center type is determined based on a seventh trained AI algorithm, wherein the seventh trained AI algorithm is configured to learn, using the training data set of molecule data structures, based on a given bond structure and on a structure of the graph data objects.

### Claim 13 (depends on 1)

The method of claim 1, wherein the likelihood value that the given atom of the plurality of assessed atoms of each of one or more graph data objects is likely the third reaction center type is determined based on an eighth trained Al algorithm, wherein the eighth trained AI algorithm is configured to learn, using the training data set of molecule data structures, based on an atom and a structure of the graph data objects.

### Claim 14 (depends on 1)

The method of claim 1, wherein the input molecule data structure is evaluated along with a plurality of input molecule data structures in a pipeline operation.

### Claim 15 (depends on 1)

The method of claim 1, wherein the plurality of input molecule data structures has an average molecule size between 20-500 atoms or rings.

### Claim 16 (depends on 1)

The method of claim 1, wherein the input molecule data structure is of an existing therapeutic molecule or a pharmaceutically active molecule.

### Claim 17 (depends on 6)

The method of claim 6, wherein the chemical substructure is iteratively attached in a ranked manner from the library of candidate substructures without any chemical rules or templates imposed for the attachment.

### Claim 18 (depends on 1)

The method of claim 1, wherein the input molecule data structure was generated by molecule optimization evaluation for drug discovery analysis.

### Claim 19

(canceled)

### Claim 20

(canceled)

### Claim 21

A system, comprising:
a processor; and
a memory operatively connected to the processor, the memory having instructions stored thereon that, when executed by the processor, cause the processor to:
receive an input molecule data structure represented as a graph data object;
evaluate one or more graph data objects of the input molecule data structure, wherein evaluating includes a set of sequences to:
determine via one or more trained AI algorithm via a training data set of molecule data structures, for one or more reaction center types, a likelihood value that a given bond of a plurality of assessed bonds or a given atom of a plurality of assessed atoms of each of one or more graph data objects is likely a reaction center for a subsequent synthesis, wherein the one or more reaction center types include:
i) a first reaction center type associated with a new chemical bond formation that is formed during a hypothetical reaction and exists in an input target molecule but not in an output reactants molecules;
ii) a second reaction center type associated with an existing chemical bond in the input molecule data structure being changed in type during a hypothetical reaction; and/or
iii) a third reaction center type associated with an atom in the input molecule data structure connected with fragments being removed during a hypothetical reaction without a new chemical bond being formed or a type of the existing chemical bond being changed,
wherein the likelihood values of the plurality of assessed bonds of each of one or more graph data objects are used in a selection of one or more reaction centers for the one or more graph data objects to generate a plurality of intermediate molecular graph data objects each having a modified chemical bond at the selected one or more reaction centers for the one or more graph data objects, wherein one or more of the plurality of intermediate molecular graph data objects with the modified chemical bond are attached to a chemical fragment at the modified chemical bond to generate a completed reactant graph, wherein the completed reactant graph is used to generate a synthetic molecule model.

### Claim 22

A non-transitory computer-readable storage medium having instructions stored thereon that, when executed by a processor, cause the processor to:
receive an input molecule data structure represented as a graph data object;
evaluate one or more graph data objects of the input molecule data structure, wherein evaluating includes a set of sequences to:
determine via one or more trained AI algorithm via a training data set of molecule data structures, for one or more reaction center types, a likelihood value that a given bond of a plurality of assessed bonds or a given atom of a plurality of assessed atoms of each of one or more graph data objects is likely a reaction center for a subsequent synthesis, wherein the one or more reaction center types include:
i) a first reaction center type associated with a new chemical bond formation that is formed during a hypothetical reaction and exists in an input target molecule but not in an output reactants molecules;
ii) a second reaction center type associated with an existing chemical bond in the input molecule data structure being changed in type during a hypothetical reaction; and/or
iii) a third reaction center type associated with an atom in the input molecule data structure connected with fragments being removed during a hypothetical reaction without a new chemical bond being formed or a type of the existing chemical bond being changed,
wherein the likelihood values of the plurality of assessed bonds of each of one or more graph data objects are used in a selection of one or more reaction centers for the one or more graph data objects to generate a plurality of intermediate molecular graph data objects each having a modified chemical bond at the selected one or more reaction centers for the one or more graph data objects, wherein one or more of the plurality of intermediate molecular graph data objects with the modified chemical bond are attached to a chemical fragment at the modified chemical bond to generate a completed reactant graph, wherein the completed reactant graph is used to generate a synthetic molecule model.
