# US 2025/0166744 A1 · System and method for predictive candidate compound discovery

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

- Number: US 2025/0166744 A1
- Document ID: US20250166744A1
- Status: Published application
- Kind: A1
- Application number: 18/832,598
- Filed: 2023-01-24
- Published: 2025-05-22
- Earliest claimed: 2022-01-24
- Assignees: Fasttrac LLC
- Inventors: Kenneth BEAN; Paul RETLEWSKI; Sujaiy SHIVAKUMAR
- Web page: https://patentgrep.com/patent/US20250166744A1

## CPC

- G16C 20/80: Data visualisation
- G16H 10/20: for electronic clinical trials or questionnaires
- G06F 3/011: {Arrangements for interaction with the human body, e.g. for user immersion in virtual reality (blind teaching G09B21/00)}
- G06F 3/017: {Gesture based interaction, e.g. based on a set of recognized hand gestures (interaction based on gestures traced on a digitiser G06F3/04883)}
- G06N 3/10: Interfaces, programming languages or software development kits, e.g. for simulating neural networks
- G06T 15/00: Three-dimensional [3D] image rendering
- G16B 15/30: Drug targeting using structural data; Docking or binding prediction
- G16B 40/20: Supervised data analysis
- G16C 20/30: Prediction of properties of chemical compounds, compositions or mixtures
- G16C 20/50: Molecular design, e.g. of drugs
- G06F 3/0482: Interaction with lists of selectable items, e.g. menus
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N 20/00: Machine learning
- G06T 2200/24: involving graphical user interfaces [GUIs]
- G06T 2210/41: Medical
- G16C 20/70: Machine learning, data mining or chemometrics
- G16C: COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G16B: BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY

## Abstract

A computing system for evaluating candidate molecules for use in candidate compound discovery is described. The system has a non-transitory computer-readable and a processor configured to receive a type of standardized data from at least a database, receive a user query, process each of the standardized data types with one or more trained machine learning models, generate an interactive environment comprising graphical representation of the candidate molecule and candidate compound based on the user query; based on a received a signal from a user, alter a configuration of the candidate compound to allow the user, using virtual reality (VR), augmented reality (AR) or both, to interact with the candidate compound and the plurality of candidate molecules and generate clinical characteristics of the candidate compound based on the further alteration by the user.

## Claims

### Claim 1

A computing system for evaluating candidate molecules for use in candidate compound discovery, the computer system comprising:
a non-transitory computer-readable memory; and
a processor configured to execute instructions stored on the non-transitory computer-readable memory which, when executed, cause the processor to:
receive a type of standardized data from at least a database, wherein the standardized data is for compounds, metals, molecules, atoms, sub-atomic particles or any combination thereof, and wherein the standardized data types comprise numerical data sets, images, graphs, text, or any combination thereof;
receive a user query, wherein the user query comprises a request for a desired attribute for the candidate compound;
process each of the standardized data types with one or more trained machine learning models, wherein the one or more machine learning models are configured to recommend at least one of a plurality of the candidate molecules to be bonded with the candidate compound based on the user query;
generate an interactive environment comprising graphical representation of the candidate molecule and candidate compound based on the user query;
based on a received a signal from a user, alter a configuration of the candidate compound to allow the user, using virtual reality (VR), augmented reality (AR) or both, to interact with the candidate compound and the plurality of candidate molecules, wherein based on a received additional signal from the user, further alter the candidate compound with another of the plurality of molecules; and
generate clinical characteristics of the candidate compound based on the further alteration by the user.

### Claim 2 (depends on 1)

The system of claim 1, wherein the instructions, when executed, further cause the processor to:
based on the type of the standardized data received, select one of the one or more machine learning models to output a trained data set, wherein:
if the standardized data is the numerical data set, execute a neural net model;
if the standardized data set is an image, execute a convolutional neural network model;
if the standardized data set is a graph, execute a multilayer perceptron model; and
If the standardized data is a text, execute a natural language processing with neural nets model.

### Claim 3 (depends on 2)

The system of claim 2, wherein the instructions, when executed, further cause the processor to:
pool the trained data sets and combine the sets; and
perform a Pareto analysis to output a predicted most accurate of the data sets.

### Claim 4 (depends on 3)

The system of claim 3, wherein the instructions, when executed, further cause the processor to:
further process the trained data sets using a recurrent neural network model to form a loop for convergence.

### Claim 5 (depends on 1)

The system of claim 1, wherein the standardized data comprises bonding angles, electron, proton and neutron configurations, melting point, toxicity, physical characteristic, chemical characteristic, atomical characteristic, biological dimensions, or any combination thereof.

### Claim 6 (depends on 1)

The system of claim 1, wherein the instructions, when executed, further cause the processor to load normalize the standardized data received.

### Claim 7 (depends on 1)

The system of claim 1, wherein the instructions, when executed, further cause the processor to:
transform the standardized data into numerical data sets; and
tag, index and assign a value to each of the data set based on the user query.

### Claim 8 (depends on 3)

The system of claim 3, wherein the instructions, when executed, further cause the processor to:
generate, using a neural net, a first user interface;
subsequent the user input, load a menu on the user interface with recommended molecules based on the user input using the neural net; and
order the recommended molecules in the menu based on a predictive success parameter using the trained data sets.

### Claim 9 (depends on 8)

The system of claim 8, wherein the instructions, when executed, further cause the processor to:
load a second layer menu on the first user interface;
order the recommended metals on the second layer menu based the predictive success parameter using the trained data sets.

### Claim 10 (depends on 9)

The system of claim 9, wherein the instructions, when executed, further cause the processor to:
define a virtual network configured to facilitate communication between at least a database and a plurality of servers having the processor and the memory in communication therewith.

### Claim 11 (depends on 1)

The system of claim 1, wherein the instructions, when executed, further cause the processor to:
When generating the interactive environment, further generate a caching layer to construct a spider graph, radar chart, or both.

### Claim 12 (depends on 1)

The system of claim 1, wherein the instructions, when executed, further cause the processor to:
receive a request at a load balancer, and evaluate listener rules in a priority order to determine which of the listener rules to apply;
select a target from a target group for listener rule to route requests to different target groups based on the content of application traffic;
execute a cloud deep learning model to run a three-dimensional rendering script to produce the interactive environment comprising the candidate molecule in atomic resolution.

### Claim 13 (depends on 12)

The system of claim 12, wherein the instructions, when executed, further cause the processor to:
create on-demand instance and a plurality of stacks for a streaming application and associate a fleet comprising a plurality of streaming instances, wherein the stack comprises an associated fleet to produce the candidate molecule and corresponding sub-atomic particles in a sub-atomic visualization.

### Claim 14

A computer-implemented method candidate compound discovery, comprising executing on a processor the steps of:
receiving a type of standardized data from at least a database, wherein the standardized data is for compounds, metals, molecules, atoms, sub-atomic particles or any combination thereof, and wherein the standardized data types comprise numerical data sets, images, graphs, text, or any combination thereof;
receiving a user query, wherein the user query comprises a request for a desired attribute for the candidate compound;
processing each of the standardized data types with one or more trained machine learning models, wherein the one or more machine learning models are configured to recommend at least one of a plurality of the candidate molecules to be bonded with the candidate compound based on the user query;
generating an interactive environment comprising graphical representation of the candidate molecule and candidate compound based on the user query;
based on a received a signal from a user, altering a configuration of the candidate compound to allow the user, using virtual reality (VR), augmented reality (AR) or both, to interact with the candidate compound and the plurality of candidate molecules, wherein based on a received additional signal from the user, further alter the candidate compound with another of the plurality of molecules; and
generating clinical characteristics of the candidate compound based on the further alteration by the user.

### Claim 15 (depends on 14)

The method of claim 14, further comprising:
based on the type of the standardized data received, selecting one of the one or more machine learning models to output a trained data set, wherein:
if the standardized data is the numerical data set, execute a neural net model;
if the standardized data set is an image, execute a convolutional neural network model;
if the standardized data set is a graph, execute a multilayer perceptron model; and
If the standardized data is a text, execute a natural language processing with neural nets model.

### Claim 16 (depends on 15)

The method of claim 15, further comprising:
pooling the trained data sets and combine the sets; and
performing a Pareto analysis to output a predicted most accurate of the data sets.

### Claim 17 (depends on 14)

The method of claim 14, further comprising:
generating, using a neural net, a first user interface;
subsequent the user input, loading a menu on the user interface with recommended molecules based on the user input using the neural net; and
ordering the recommended molecules in the menu based a predictive success parameter using the trained data sets.

### Claim 18 (depends on 17)

The method of claim 17 further comprising defining a virtual network configured to facilitate communication between at least one database and a plurality of servers having the processor and the memory in communication therewith.

### Claim 19 (depends on 14)

The method of claim 14, further comprising:
receiving a request at a load balancer, and evaluate listener rules in a priority order to determine which of the listener rules to apply;
selecting a target from a target group for listener rule to route requests to different target groups based on the content of application traffic;
executing a cloud deep learning model to run a three-dimensional rendering script to produce the interactive environment comprising the candidate molecule in atomic resolution.

### Claim 20

A non-transitory computer-readable medium having stored thereon program instructions that, upon execution by a computing device, cause the computing device to perform operations comprising:
receive a type of standardized data from at least a database, wherein the standardized data is for compounds, metals, molecules, atoms, sub-atomic particles or any combination thereof, and wherein the standardized data types comprise numerical data sets, images, graphs, text, or any combination thereof;
receive a user query, wherein the user query comprises a request for a desired attribute for the candidate compound;
process each of the standardized data types with one or more trained machine learning models, wherein the one or more machine learning models are configured to recommend at least one of a plurality of the candidate molecules to be bonded with the candidate compound based on the user query;
generate an interactive environment comprising graphical representation of the candidate molecule and candidate compound based on the user query;
based on a received a signal from a user, alter a configuration of the candidate compound to allow the user, using virtual reality (VR), augmented reality (AR) or both, to interact with the candidate compound and the plurality of candidate molecules, wherein based on a received additional signal from the user, further alter the candidate compound with another of the plurality of molecules; and
generate clinical characteristics of the candidate compound based on the further alteration by the user.
