# US 2021/0343370 A1 · Cross-variant polygenic predictive data analysis

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

- Number: US 2021/0343370 A1
- Document ID: US20210343370A1
- Status: Published application
- Kind: A1
- Application number: 16/863,362
- Filed: 2020-04-30
- Published: 2021-11-04
- Earliest claimed: 2020-04-30
- Assignees: Optum Services (Ireland) Limited
- Applicants: Optum Services (Ireland) Limited.
- Inventors: Kenneth Bryan; Megan O'Brien; David S. Monaghan; Chirag Chadha
- Web page: https://patentgrep.com/patent/US20210343370A1

## CPC

- G16B 20/40: Population genetics; Linkage disequilibrium
- G06F 18/2113: by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- G16B 10/00: ICT specially adapted for evolutionary bioinformatics, e.g. phylogenetic tree construction or analysis
- G16B 20/20: Allele or variant detection, e.g. single nucleotide polymorphism [SNP] detection
- G16B 30/00: ICT specially adapted for sequence analysis involving nucleotides or amino acids
- G16B 40/00: ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
- G16B 40/20: Supervised data analysis
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- G16H 50/30: for calculating health indices; for individual health risk assessment
- G06N 3/045: Combinations of networks
- G16B: BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

There is a need for more effective and efficient predictive data analysis solutions for processing genetic sequencing data. This need can be addressed by, for example, techniques for performing predictive data analysis based on genetic sequences that utilize at least one of cross-variant polygenic risk modeling using genetic risk profiles, cross-variant polygenic risk modeling using functional genetic risk profiles, per-condition polygenic clustering operations, cross-condition polygenic predictive inferences, and cross-condition polygenic diagnoses.

## Claims

### Claim 1

A computer-implemented method for cross-variant polygenic risk modeling for a target individual in relation to a target medical condition, the computer-implemented method comprising:
generating, for each genetic variant of a plurality of genetic variants that is associated with a chromosome of a plurality of chromosomes, a per-variant genetic risk score of a plurality of per-variant genetic risk scores based on: (i) a per-variant risk probability value for the genetic variant in relation to the target medical condition that is selected from a plurality of per-variant risk probability values for the plurality of genetic variants, and (ii) a per-variant allele count for the genetic variant in relation to the target individual that is selected from a plurality of per-variant allele counts for the plurality of genetic variants;
generating a genetic risk profile for the plurality of genetic variants, wherein: (i) the genetic risk profile comprises one or more per-chromosome profile segments each associated with a chromosome of the plurality of chromosomes, and (ii) each per-chromosome profile segment of the one or more per-chromosome profile segments that is associated with a respective chromosome of the plurality of chromosomes comprises each per-variant genetic risk score for a genetic variant in a subset of the plurality of genetic variants that is associated with the respective chromosome; and
performing one or more prediction-based actions based on the genetic risk profile.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein generating the plurality of per-variant risk probability values comprises:
identifying one or more training genetic sequences, wherein each training genetic sequence of the one or more training genetic sequences is associated with a training observational label in relation to the target medical condition;
generating, based on the one or more training genetic sequences, a plurality of per-candidate-variant risk probability values, wherein the plurality of per-candidate-variant risk probability values comprise a per-candidate-variant risk probability value for each candidate genetic variant of a plurality of candidate genetic variants;
determining a selected subset of the plurality of candidate genetic variants as the plurality of genetic variants; and
generating the plurality of per-variant risk probability values based on each selected per-candidate-variant risk probability value for a candidate genetic variant that is in the selected sub set.

### Claim 3 (depends on 2)

The computer-implemented method of claim 2, wherein each per-candidate-variant risk probability value for a candidate genetic variant of the plurality of candidate genetic variants is a log of odds ratio of the candidate genetic variant in relation to the target medical condition as determined in accordance with the one or more training genetic sequences.

### Claim 4 (depends on 2)

The computer-implemented method of claim 2, wherein generating the plurality of per-variant risk probability values based on each selected per-candidate-variant risk probability value comprises:
identifying one or more testing genetic sequences, wherein each testing genetic sequence of the one or more testing genetic sequences is associated with a testing observational label in relation to the target medical condition;
determining, based on processing the one or more training genetic sequences in accordance with each selected per-candidate-variant risk probability value, a validation determination; and
in response to determining that the validation determination indicates a positive value, adopting each selected per-candidate-variant risk probability value as a corresponding per-variant risk probability value for a corresponding genetic variant in the plurality of genetic variants.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein performing the one or more prediction-based actions based on the genetic risk profile comprises:
generating a chromosome-grouped predictive output interface comprising a chromosome-grouped predictive output graph, wherein a first coordinate of the chromosome-grouped predictive output graph comprises descriptions of the plurality of genetic variants in accordance with the one or more per-chromosome profile segments and a second coordinate of the chromosome-grouped predictive output graph comprises a range of the plurality of per-variant genetic risk scores.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein:
the plurality of per-variant allele counts is determined based on a plurality of per-variant zygosity values, and
each per-variant zygosity value of the plurality of per-variant zygosity values is associated with a genetic variant of the plurality of genetic variants.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of genetic variants comprise one or more static genetic variants determined based on predictive domain data associated with the target medical condition.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of genetic variants comprise one or more dynamic genetic variants whose respective p-values with respect to the target medical condition exceed a threshold risk probability value.

### Claim 9

An apparatus for cross-variant polygenic risk modeling for a target individual in relation to a target medical condition, the apparatus comprising at least one processor and at least one memory including program code, the at least one memory and the program code configured to, with the processor, cause the apparatus to at least:
generate, for each genetic variant of a plurality of genetic variants that is associated with a chromosome of a plurality of chromosomes, a per-variant genetic risk score of a plurality of per-variant genetic risk scores based on: (i) a per-variant risk probability value for the genetic variant in relation to the target medical condition that is selected from a plurality of per-variant risk probability values for the plurality of genetic variants, and (ii) a per-variant allele count for the genetic variant in relation to the target individual that is selected from a plurality of per-variant allele counts for the plurality of genetic variants;
generate a genetic risk profile for the plurality of genetic variants, wherein: (i) the genetic risk profile comprises one or more per-chromosome profile segments each associated with a chromosome of the plurality of chromosomes, and (ii) each per-chromosome profile segment of the one or more per-chromosome profile segments that is associated with a respective chromosome of the plurality of chromosomes comprises each per-variant genetic risk score for a genetic variant in a subset of the plurality of genetic variants that is associated with the respective chromosome; and
performing one or more prediction-based actions based on the genetic risk profile.

### Claim 10 (depends on 9)

The apparatus of claim 9, wherein generating the plurality of per-variant risk probability values comprises:
identifying one or more training genetic sequences, wherein each training genetic sequence of the one or more training genetic sequences is associated with a training observational label in relation to the target medical condition;
generating, based on the one or more training genetic sequences, a plurality of per-candidate-variant risk probability values, wherein the plurality of per-candidate-variant risk probability values comprise a per-candidate-variant risk probability value for each candidate genetic variant of a plurality of candidate genetic variants;
determining a selected subset of the plurality of candidate genetic variants as the plurality of genetic variants; and
generating the plurality of per-variant risk probability values based on each selected per-candidate-variant risk probability value for a candidate genetic variant that is in the selected subset.

### Claim 11 (depends on 10)

The apparatus of claim 10, wherein each per-candidate-variant risk probability value for a candidate genetic variant of the plurality of candidate genetic variants is a log of odds ratio of the candidate genetic variant in relation to the target medical condition as determined in accordance with the one or more training genetic sequences.

### Claim 12 (depends on 10)

The apparatus of claim 10, wherein generating the plurality of per-variant risk probability values based on each selected per-candidate-variant risk probability value comprises:
identifying one or more testing genetic sequences, wherein each testing genetic sequence of the one or more testing genetic sequences is associated with a testing observational label in relation to the target medical condition;
determining, based on processing the one or more training genetic sequences in accordance with each selected per-candidate-variant risk probability value, a validation determination; and
in response to determining that the validation determination indicates a positive value, adopting each selected per-candidate-variant risk probability value as a corresponding per-variant risk probability value for a corresponding genetic variant in the plurality of genetic variants.

### Claim 13 (depends on 9)

The apparatus of claim 9, wherein performing the one or more prediction-based actions based on the genetic risk profile comprises:
generating a chromosome-grouped predictive output interface comprising a chromosome-grouped predictive output graph, wherein a first coordinate of the chromosome-grouped predictive output graph comprises descriptions of the plurality of genetic variants in accordance with the one or more per-chromosome profile segments and a second coordinate of the chromosome-grouped predictive output graph comprises a range of the plurality of per-variant genetic risk scores.

### Claim 14 (depends on 9)

The apparatus of claim 9, wherein:
the plurality of per-variant allele counts is determined based on a plurality of per-variant zygosity values, and
each per-variant zygosity value of the plurality of per-variant zygosity values is associated with a genetic variant of the plurality of genetic variants.

### Claim 15 (depends on 9)

The apparatus of claim 9, wherein the plurality of genetic variants comprise one or more static genetic variants determined based on predictive domain data associated with the target medical condition.

### Claim 16 (depends on 9)

The apparatus of claim 9, wherein the plurality of genetic variants comprise one or more dynamic genetic variants whose respective p-values with respect to the target medical condition exceed a threshold risk probability value.

### Claim 17

A computer program product for cross-variant polygenic risk modeling for a target individual in relation to a target medical condition, the computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions configured to:
generate, for each genetic variant of a plurality of genetic variants that is associated with a chromosome of a plurality of chromosomes, a per-variant genetic risk score of a plurality of per-variant genetic risk scores based on: (i) a per-variant risk probability value for the genetic variant in relation to the target medical condition that is selected from a plurality of per-variant risk probability values for the plurality of genetic variants, and (ii) a per-variant allele count for the genetic variant in relation to the target individual that is selected from a plurality of per-variant allele counts for the plurality of genetic variants;
generate a genetic risk profile for the plurality of genetic variants, wherein: (i) the genetic risk profile comprises one or more per-chromosome profile segments each associated with a chromosome of the plurality of chromosomes, and (ii) each per-chromosome profile segment of the one or more per-chromosome profile segments that is associated with a respective chromosome of the plurality of chromosomes comprises each per-variant genetic risk score for a genetic variant in a subset of the plurality of genetic variants that is associated with the respective chromosome; and
performing one or more prediction-based actions based on the genetic risk profile.

### Claim 18 (depends on 17)

The computer program product of claim 17, wherein generating the plurality of per-variant risk probability values comprises:
identifying one or more training genetic sequences, wherein each training genetic sequence of the one or more training genetic sequences is associated with a training observational label in relation to the target medical condition;
generating, based on the one or more training genetic sequences, a plurality of per-candidate-variant risk probability values, wherein the plurality of per-candidate-variant risk probability values comprise a per-candidate-variant risk probability value for each candidate genetic variant of a plurality of candidate genetic variants;
determining a selected subset of the plurality of candidate genetic variants as the plurality of genetic variants; and
generating the plurality of per-variant risk probability values based on each selected per-candidate-variant risk probability value for a candidate genetic variant that is in the selected subset.

### Claim 19 (depends on 18)

The computer program product of claim 18, wherein each per-candidate-variant risk probability value for a candidate genetic variant of the plurality of candidate genetic variants is a log of odds ratio of the candidate genetic variant in relation to the target medical condition as determined in accordance with the one or more training genetic sequences.

### Claim 20 (depends on 18)

The computer program product of claim 18, wherein generating the plurality of per-variant risk probability values based on each selected per-candidate-variant risk probability value comprises:
identifying one or more testing genetic sequences, wherein each testing genetic sequence of the one or more testing genetic sequences is associated with a testing observational label in relation to the target medical condition;
determining, based on processing the one or more training genetic sequences in accordance with each selected per-candidate-variant risk probability value, a validation determination; and
in response to determining that the validation determination indicates a positive value, adopting each selected per-candidate-variant risk probability value as a corresponding per-variant risk probability value for a corresponding genetic variant in the plurality of genetic variants.