IP Library › Granted Patent US 11,735,292
Granted Patent B2
US 11,735,292 · App. 16/057,673 · Granted Aug 22, 2023

Intelligent personalized chemical synthesis planning

Inventors: Akihiro Kishimoto (Castleknock, IE); Adi I. Botea (Dublin, IE); Beat Buesser (Ashtown, IE); Bei Chen (Blanchardstown, IE); Takahiko Koyama (Scarsdale, NY)
Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
G16C20/10G06F16/90335G06N5/025G06F17/13
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,735,292
App. No.
16/057,673
Granted
Aug 22, 2023
Kind
B2
Abstract

Embodiments for intelligent discovery of preferred chemical reaction pathways to a synthesized target compound by a processor. Target compound data of a target compound a user intends to synthesize may be received. A preference score for one or more chemical reaction pathways to the target compound may be calculated based on one or more chemical reactions. One or more preferred chemical reaction pathways to the targeted compound may be determined using a modified retrosynthetic analysis according to the preference score for one or more chemical reactions.

Claims (45)

1. A method for intelligent discovery of preferred chemical reaction pathways to a synthesized target compound in a computing environment by a processor, comprising:

receiving, by the processor, target compound data of a target compound a user intends to synthesize;

calculating, by the processor using at least one of a node graph searching operation and a machine learning operation, a preference score for one or more chemical reaction pathways to the target compound based on one or more chemical reactions, wherein the preference score is personalized to a skill level of the user and calculated based on detected preferences of the user, the detected preferences inclusive of starting materials and equipment available to the user to synthesize the target compound;

determining, by the processor as an output of the at least one of the node graph searching operation and the machine learning operation, one or more preferred chemical reaction pathways to synthesize the target compound using a modified retrosynthetic analysis according to the preference score for one or more chemical reactions, wherein, given input of the target compound, the determination of precursor compounds of the one or more preferred chemical reaction pathways to synthesize the target compound are returned as the output in one reaction step; and

displaying, by the processor via an interactive graphical user interface (GUI), the one or more preferred chemical reaction pathways to the target compound listed in an order based on the preference score for the one or more chemical reactions.

2. The method of claim 1 , further including:

receiving a list of chemical reaction pathways preferred by the user; or

receiving a list of chemical reaction pathways non-preferred by the user.

3. The method of claim 1 , further including:

receiving one or more reaction rules from a reaction rule database for determining the one or more preferred chemical reaction pathways; or

receiving one or more reaction types describing a type of reaction for determining the one or more preferred chemical reaction pathways.

4. The method of claim 1 , further including:

collecting feedback relating to the or more preferred chemical reaction pathways; or

searching for one or more alternative preferred chemical reaction pathways according to the feedback.

5. The method of claim 1 , further including initializing a machine learning operation to predict one or more reactants from one or more products and learn one or more starting materials required for synthesizing the target compound.

6. A system for intelligent discovery of preferred chemical reaction pathways to a synthesized target compound, comprising:

one or more computers with executable instructions that when executed by a processor cause the system to:

receive, by the processor, target compound data of a target compound a user intends to synthesize;

calculate, by the processor using at least one of a node graph searching operation and a machine learning operation, a preference score for one or more chemical reaction pathways to the target compound based on one or more chemical reactions, wherein the preference score is personalized to a skill level of the user and calculated based on detected preferences of the user, the detected preferences inclusive of starting materials and equipment available to the user to synthesize the target compound;

determine, by the processor as an output of the at least one of the node graph searching operation and the machine learning operation, one or more preferred chemical reaction pathways to synthesize the target compound using a modified retrosynthetic analysis according to the preference score for one or more chemical reactions, wherein, given input of the target compound, the determination of precursor compounds of the one or more preferred chemical reaction pathways to synthesize the target compound are returned as the output in one reaction step; and

display, by the processor via an interactive graphical user interface (GUI), the one or more preferred chemical reaction pathways to the target compound listed in an order based on the preference score for the one or more chemical reactions.

7. The system of claim 6 , wherein the executable instructions further:

receive a list of chemical reaction pathways preferred by the user; or

receive a list of chemical reaction pathways non-preferred by the user.

8. The system of claim 6 , wherein the executable instructions further:

receive one or more reaction rules from a reaction rule database for determining the one or more preferred chemical reaction pathways; and

receive one or more reaction types describing a type of reaction for determining the one or more preferred chemical reaction pathways.

9. The system of claim 6 , wherein the executable instructions further:

collect feedback relating to the or more preferred chemical reaction pathways; or

search for one or more alternative preferred chemical reaction pathways according to the feedback.

10. The system of claim 6 , wherein the executable instructions further initialize a machine learning operation to predict one or more reactants from one or more products and learn one or more starting materials required for synthesizing the target compound.

11. A computer program product for intelligent discovery of preferred chemical reaction pathways to a synthesized target compound by a processor, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:

an executable portion that receives, by the processor, target compound data of a target compound a user intends to synthesize;

an executable portion that calculates, by the processor using at least one of a node graph searching operation and a machine learning operation, a preference score for one or more chemical reaction pathways to the target compound based on one or more chemical reactions, wherein the preference score is personalized to a skill level of the user and calculated based on detected preferences of the user, the detected preferences inclusive of starting materials and equipment available to the user to synthesize the target compound;

an executable portion that determines, by the processor as an output of the at least one of the node graph searching operation and the machine learning operation, one or more preferred chemical reaction pathways to synthesize the target compound using a modified retrosynthetic analysis according to the preference score for one or more chemical reactions, wherein, given input of the target compound, the determination of precursor compounds of the one or more preferred chemical reaction pathways to synthesize the target compound are returned as the output in one reaction step; and

an executable portion that displays, by the processor via an interactive graphical user interface (GUI), the one or more preferred chemical reaction pathways to the target compound listed in an order based on the preference score for the one or more chemical reactions.

12. The computer program product of claim 11 , wherein the executable portion that:

receives a list of chemical reaction pathways preferred by the user;

receives a list of chemical reaction pathways non-preferred by the user;

receives one or more reaction rules from a reaction rule database for determining the one or more preferred chemical reaction pathways; or

receive one or more reaction types describing a type of reaction for determining the one or more preferred chemical reaction pathways.

13. The computer program product of claim 11 , wherein the executable portion that:

collects feedback relating to the or more preferred chemical reaction pathways; or

searches for one or more alternative preferred chemical reaction pathways according to the feedback.

14. The computer program product of claim 11 , wherein the executable portion that initializes a machine learning operation to predict one or more reactants from one or more products and learn one or more starting materials required for synthesizing the target compound.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 7, 2018
From: KISHIMOTO, AKIHIRO; BOTEA, ADI I.; BUESSER, BEAT; CHEN, BEI; KOYAMA, TAKAHIKO
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 046577/0224 →
Continuity (1)
Related Publication 20200050947A1 · Feb 13, 2020
Cited By (5)
US 12,368,503 US 12,587,274 US 12,603,701 US 12,627,372 US 12,664,813