Genotype Phenotype Prediction Using Hidden Markov Models

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current genotype-phenotype association studies face limitations such as massive genotyping requirements, insufficient statistical power, and non-causal correlations, particularly in identifying associations with rare variants and combinatorial genotypes, which restrict their predictive capabilities.

Innovation Solution

The method involves accessing genomic data from a subject and a background population, using Hidden Markov Models and spectral clustering to identify relevant genetic variants associated with phenotypes, determining the strength-of-effect and ontological terms, and generating reports on phenotypic predictions, allowing for causal relationship determination and prediction of rare or unique variants.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional genotype-phenotype association studies are used, then statistical power is insufficient, but requiring massive genotyping of background samples increases complexity and cost

Engineering Contradiction:
Improvestatistical powerVSAvoidgenotyping requirements
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and focuses only on the most informative genetic variants for phenotype prediction, rather than analyzing the entire genome. By identifying and selecting specific variants with high predictive value, the method achieves high statistical power with minimal genotyping requirements, eliminating the need for massive background sample analysis

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the analytical parameter from comprehensive genome-wide association to targeted variant effect prediction. By using in silico prediction tools to assess variant impact on protein function and phenotype, the method transforms the problem from statistical correlation in large populations to functional prediction from limited data

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If comprehensive genome analysis is performed to identify all variants, then coverage is complete, but computational complexity and data processing requirements increase significantly

Engineering Contradiction:
Improvevariant identification accuracyVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the most relevant variants for phenotype prediction by using in silico tools to identify variants with predicted functional impact. This selective extraction eliminates the need to process entire genomes while maintaining high accuracy for the variants that matter most for phenotype determination

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the genome analysis into focused assessments of individual variant effects rather than comprehensive population-wide association studies. By evaluating each variant's predicted impact independently using computational tools, the method reduces computational complexity while maintaining precision

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If population-specific background samples are used to improve relevance, then predictive accuracy for specific populations improves, but the method loses generality and requires multiple population-specific datasets

Engineering Contradiction:
Improvepredictive accuracyVSAvoidpopulation generality
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal prediction system that functions across all populations by using in silico variant effect prediction rather than population-specific statistical associations. The method evaluates the functional impact of variants independently of population frequency, making it universally applicable while maintaining accuracy for rare and population-specific variants

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Quantity of substance

If traditional association studies are used, then common variants can be identified, but rare variants and combinatorial genotypes cannot be effectively detected

Engineering Contradiction:
Improvenumber of identifiable variantsVSAvoidcausal relationship determination
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent changes from statistical association measurement to functional impact prediction as the key parameter. By using in silico tools to predict how variants affect protein function and phenotype, the method can identify rare variants and combinatorial genotypes that traditional frequency-based association studies miss, while establishing causal relationships through functional prediction

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3405897B1Determining phenotype from genotype
Publication Date: 2024.12.04 OUTSEE LTD
  • EP3405897B1 patent drawingFigure 1A
  • EP3405897B1 patent drawingFigure 1B
  • EP3405897B1 patent drawingFigure 1C

AI summary

Examining genomic information for a specific variance, while useful, provides limited information. Accordingly, systems and methods are provided to analyze a subject's genome against a background population. Outlier variances that are known ontological terms having at least a threshold strength-of-effect are then determined between each member. As a benefit, the subject's outlier variances, which may be further ranked in terms of relationship to a known phenotype, may be identified for an entirety or portion of a genomic sequence.