# US 2025/0157595 A1 · Utilizing compound-protein machine learning representations to generate bioactivity predictions

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

- Number: US 2025/0157595 A1
- Document ID: US20250157595A1
- Status: Published application
- Kind: A1
- Application number: 18/505,728
- Filed: 2023-11-09
- Published: 2025-05-15
- Earliest claimed: 2023-11-09
- Assignees: Recursion Pharmaceuticals, Inc.
- Applicants: Recursion Pharmaceuticals, Inc.
- Inventors: Seyed Ali Madani Tonekaboni; Daniella Fiora Lato; Stephen Scott MacKinnon
- Web page: https://patentgrep.com/patent/US20250157595A1

## CPC

- G16C 20/70: Machine learning, data mining or chemometrics
- G16B 15/30: Drug targeting using structural data; Docking or binding prediction
- G16B 40/20: Supervised data analysis
- 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

The present disclosure relates to systems, non-transitory computer-readable media, and methods that utilizing compound-protein machine learning representations to generate target results. For example, the disclosed systems can utilize a compound-protein interaction machine learning model to generate a compound-protein machine learning representation for compound protein pairs. The disclosed systems can utilize the compound-protein machine learning representation to train and utilize other target machine learning models in generating predicted bioactivity results. For example, the disclosed systems train a target machine learning model from compound-protein machine learning representations to generate ADMET predictions and/or biological perturbation program predictions. Furthermore, the disclosed systems can utilize one or more explainability models in conjunction with target machine learning models trained based on compound-protein machine learning representations to identify proteins that contribute to predicted bioactivity results.

## Claims

### Claim 1

A computer-implemented method comprising:
identifying a plurality of compound-protein pairs comprising a training compound matched to a plurality of proteins;
generating, utilizing a compound-protein interaction machine learning model from the plurality of compound-protein pairs, a plurality of binding scores between the training compound and the plurality of proteins;
generating a compound-protein machine learning representation comprising the plurality of binding scores between the training compound and the plurality of proteins; and
iteratively training a target machine learning model by, for a first training iteration:
generating a predicted bioactivity result for the training compound, from the compound-protein machine learning representation comprising the plurality of binding scores between the training compound and the plurality of proteins, utilizing the target machine learning model; and
modifying parameters of the target machine learning model by comparing the predicted bioactivity result to a ground truth bioactivity result corresponding to the training compound.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein generating the plurality of binding scores comprises:
generating, utilizing the compound-protein interaction machine learning model, a first binding score of the plurality of binding scores indicating a first binding likelihood for a first compound-protein pair, the first compound-protein pair comprising the training compound and a first protein; and
generating, utilizing the compound-protein interaction machine learning model, a second binding score of the plurality of binding scores indicating a second binding likelihood for a second compound-protein pair, the second compound-protein pair comprising the training compound and a second protein.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the compound-protein interaction machine learning model comprises a compound-protein interaction neural network having parameters trained to generate binding scores indicating probabilities that compounds will bind to protein pockets of proteins.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, further comprising:
generating, utilizing the compound-protein interaction machine learning model, an additional compound-protein machine learning representation from an additional plurality of binding scores between an additional training compound and the plurality of proteins; and
iteratively training the target machine learning model by, for a second training iteration:
generating an additional predicted bioactivity result for the additional training compound from the additional compound-protein machine learning representation utilizing the target machine learning model; and
modifying parameters of the target machine learning model by comparing the additional predicted bioactivity result to an additional ground truth bioactivity result corresponding to the additional training compound.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein:
the predicted bioactivity result comprises an ADMET prediction, the ADMET prediction comprising a bioactivity prediction corresponding to absorption, distribution, metabolism, excretion, or toxicity; and
training the target machine learning model comprises training the target machine learning model to generate ADMET predictions by:
generating, from the compound-protein machine learning representation utilizing the target machine learning model, the ADMET prediction for the training compound; and
modifying the parameters of the target machine learning model by comparing the ADMET prediction to the ground truth bioactivity result, the ground truth bioactivity result comprising a measured ADMET result for the training compound.

### Claim 6 (depends on 4)

The computer-implemented method of claim 4, wherein the predicted bioactivity result comprises a biological perturbation program prediction and training the target machine learning model comprises training the target machine learning model to generate biological perturbation program predictions utilizing a training dataset by, for a biological perturbation program corresponding to identifying compounds demonstrating a target biological activity:
generating, from the compound-protein machine learning representation utilizing the target machine learning model, the biological perturbation program prediction for the training compound; and
modifying the parameters of the target machine learning model by comparing the biological perturbation program prediction with a ground truth perturbation program result.

### Claim 7 (depends on 6)

The computer-implemented method of claim 6, further comprising generating the training dataset by:
generating measures of similarity between a target gene of the biological perturbation program and datapoints of the training dataset, wherein the measures of similarity are based on at least one of phenomic data, transcriptomic data, metabolomic data, or proteomic data; and
generating the training dataset by filtering datapoints based on the measures of similarity between the datapoints and the target gene.

### Claim 8 (depends on 7)

The computer-implemented method of claim 7, wherein the phenomic data comprises phenomic image embeddings and further comprising generating the training dataset by:
identifying phenomic digital images of cell perturbations;
generating, utilizing a machine learning model, phenomic image embeddings from the phenomic digital images; and
generating the training dataset by filtering datapoints based on pheno-similarity measures of the phenomic image embeddings relative to a target gene.

### Claim 9 (depends on 6)

The computer-implemented method of claim 6, further comprising generating the training dataset by:
generating, utilizing a clustering model, clusters from a dataset utilizing chemical fingerprints of compounds; and
splitting the dataset into the training dataset and a testing data set based on the clusters.

### Claim 10

A system comprising:
at least one processor; and
at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
identifying a plurality of compound-protein pairs comprising a training compound matched to a plurality of proteins;
generate, utilizing a compound-protein interaction machine learning model from the plurality of compound-protein pairs, a plurality of binding scores between the training compound and the plurality of proteins;
generate a compound-protein machine learning representation comprising from the plurality of binding scores between the training compound and the plurality of proteins; and
iteratively train a target machine learning model by, for a first training iteration:
generating a predicted bioactivity result for the training compound, from the compound-protein machine learning representation comprising the plurality of binding scores between the training compound and the plurality of proteins, utilizing the target machine learning model; and
modifying parameters of the target machine learning model by comparing the predicted bioactivity result to a ground truth bioactivity result corresponding to the training compound.

### Claim 11 (depends on 10)

The system of claim 10, further comprising instructions that, when executed by the at least one processor, cause the system to generate the plurality of binding scores by:
generating, utilizing the compound-protein interaction machine learning model, a first binding score of the plurality of binding scores indicating a first binding likelihood for a first compound-protein pair, the first compound-protein pair further comprising the training compound and a first protein; and
generating, utilizing the compound-protein interaction machine learning model, a second binding score of the plurality of binding scores indicating a second binding likelihood for a second compound-protein pair, the second compound-protein pair further comprising the training compound and a second protein.

### Claim 12 (depends on 10)

The system of claim 10, further comprising instructions that, when executed by the at least one processor, cause the system to generate the compound-protein machine learning representation by:
determining machine learning protein confidence scores indicating a measure of confidence of the compound-protein interaction machine learning model in generating binding predictions for the plurality of proteins; and
filtering one or more features based on the machine learning protein confidence scores to generate the compound-protein machine learning representation.

### Claim 13 (depends on 10)

The system of claim 10, further comprising instructions that, when executed by the at least one processor, cause the system to:
generate, utilizing the compound-protein interaction machine learning model, an additional compound-protein machine learning representation from an additional plurality of binding scores between an additional training compound and the plurality of proteins; and
iteratively train the target machine learning model by, for a second training iteration:
generating an additional predicted bioactivity result for the additional training compound from the additional compound-protein machine learning representation utilizing the target machine learning model; and
modifying parameters of the target machine learning model by comparing the additional predicted bioactivity result to an additional ground truth bioactivity result corresponding to the additional training compound.

### Claim 14 (depends on 10)

The system of claim 10:
wherein the predicted bioactivity result comprises an ADMET prediction, the ADMET prediction comprising a bioactivity prediction corresponding to absorption, distribution, metabolism, excretion, or toxicity; and
further comprising instructions that, when executed by the at least one processor, cause the system to train the target machine learning model to generate ADMET predictions for a first training iteration by:
generating, from the compound-protein machine learning representation utilizing the target machine learning model, the ADMET prediction for the training compound; and
modifying the parameters of the target machine learning model by comparing the ADMET prediction to the ground truth bioactivity result, the ground truth bioactivity result comprising a measured ADMET result for the training compound.

### Claim 15 (depends on 13)

The system of claim 13, wherein the predicted bioactivity result comprises a biological perturbation program prediction and further comprising instructions that, when executed by the at least one processor, cause the system to train the target machine learning model to generate biological perturbation program predictions utilizing a training dataset by, for a biological perturbation program corresponding to identifying demonstrating a target biological activity:
generating, from the compound-protein machine learning representation utilizing the target machine learning model, the biological perturbation program prediction for the training compound; and
modifying the parameters of the target machine learning model by comparing the biological perturbation program prediction with a ground truth biological perturbation program result.

### Claim 16 (depends on 15)

The system of claim 15, further comprising instructions that, when executed by the at least one processor, cause the system to generate the training dataset by:
identifying phenomic digital images of cell perturbations;
generating, utilizing a machine learning model, phenomic image embeddings from the phenomic digital images; and
generating the training dataset by filtering datapoints based on a measure of similarity of the phenomic image embeddings relative to the target biological activity.

### Claim 17

A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computing device to:
identify a plurality of compound-protein pairs comprising a training compound matched to a plurality of proteins;
generate, utilizing a compound-protein interaction machine learning model from the plurality of compound-protein pairs, a plurality of binding scores between the training compound and the plurality of proteins;
generate a compound-protein machine learning representation comprising the plurality of binding scores between the training compound and the plurality of proteins; and
iteratively train a target machine learning model by, for a first training iteration:
generating a predicted bioactivity result for the training compound, from the compound-protein machine learning representation comprising the plurality of binding scores between the training compound and the plurality of proteins, utilizing the target machine learning model; and
modifying parameters of the target machine learning model by comparing the predicted bioactivity result to a ground truth bioactivity result corresponding to the training compound.

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, further comprising instructions that, when executed by the at least one processor, cause the computing device to generate the compound-protein machine learning representation by:
determining machine learning protein confidence scores indicating a measure of confidence of the compound-protein interaction machine learning model in generating binding predictions for the plurality of proteins; and
filtering one or more features based on the machine learning protein confidence scores to generate the compound-protein machine learning representation.

### Claim 19 (depends on 17)

The non-transitory computer-readable medium of claim 17:
wherein the predicted bioactivity result comprises an ADMET prediction, the ADMET prediction comprising a bioactivity prediction corresponding to absorption, distribution, metabolism, excretion, or toxicity; and
further comprising instructions that, when executed by the at least one processor, cause the computing device to train the target machine learning model to generate ADMET predictions for a first training iteration by:
generating, from the compound-protein machine learning representation utilizing the target machine learning model, the ADMET prediction for the training compound; and
modifying the parameters of the target machine learning model by comparing the ADMET prediction to the ground truth bioactivity result, the ground truth bioactivity result comprising a measured ADMET result for the training compound.

### Claim 20 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the predicted bioactivity result comprises a biological perturbation program prediction and further comprising instructions that, when executed by the at least one processor, cause the computing device to train the target machine learning model to generate biological perturbation program predictions utilizing a training dataset by, for a biological perturbation program corresponding to identifying compounds demonstrating a target biological activity:
generating, from the compound-protein machine learning representation utilizing the target machine learning model, the biological perturbation program prediction for a training compound; and
modifying the parameters of the target machine learning model by comparing the biological perturbation program prediction with a ground truth biological perturbation program result.
