ML Classifier for Clinical Variant Pathogenicity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing genetic diseases using genomic sequencing face challenges in accurately assessing the pathogenicity of variants of unknown significance (VUS), leading to low diagnostic yields due to the inability to reliably classify these variants as pathogenic or benign, which are prevalent in clinical reports.
Innovation Solution
The development of classifier models using machine learning systems that integrate molecular modeling with in vivo validation in animal models to predict the pathogenicity of clinical variants by analyzing phenotype and transcriptome features from transgenic organisms, enabling the classification of VUS into pathogenic or benign categories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If genomic sequencing is used to identify clinical variants, then the ability to detect genetic variations is improved, but the ability to accurately classify variants of unknown significance (VUS) as pathogenic or benign deteriorates
Solution Approach 1:
The patent introduces transgenic animal models (nematodes and zebrafish) as intermediary systems between genomic sequencing and variant classification. These models express human clinical variants and exhibit phenotypes that serve as mediators to infer pathogenicity, bridging the gap between sequence data and functional interpretation
Solution Approach 2:
The patent replaces traditional biochemical assays and cell culture systems with in vivo animal model systems. This substitution enables the study of transcellular pathogenicities and complex organismal phenotypes that cannot be captured in vitro, improving classification reliability
2Reliability
If transgenic animal models are used to assess functional consequence of missense changes, then the reliability of pathogenicity assessment is improved, but the complexity of the diagnostic system increases
Solution Approach 1:
The patent segments the diagnostic workflow into distinct modules: genomic sequencing, transgenic model generation, phenotypic assessment, and classifier model application. Each module can be independently optimized and validated, managing overall system complexity while maintaining high reliability
Solution Approach 2:
The patent changes the biological context parameter from in vitro to in vivo systems. By using transgenic nematodes and zebrafish with intact organismal biology, the system captures complex pathogenic mechanisms while standardizing phenotypic measurements for computational analysis
3Measurement precision
If a panel of multiple phenotype and transcriptome features is analyzed, then the accuracy of VUS classification is improved, but the amount of data processing and computational resources required increases
Solution Approach 1:
The patent performs preliminary action by pre-training classifier models using extensive training data from transgenic animal models with known pathogenicity. These pre-trained models can then rapidly classify new VUS with high accuracy without requiring extensive real-time computational resources
Solution Approach 2:
The patent creates computational copies of biological systems through classifier models that replicate the decision-making process of expert curators. These models process multiple phenotype and transcriptome features simultaneously, achieving high classification accuracy while reducing manual analysis burden
Data Source
AI summary
Disclosed herein are classifier models, computer implemented systems, machine learning systems and methods thereof for classifying clinical variants of unknown or uncertain significance into a pathogenicity category using measured phenotype features extracted from phenotype assays of transgenic organism expressing the human clinical variant. Embodiments of the present invention relate generally to methods for generating classifier models using machine learning and use of those classifier models to predict the pathogenicity of a clinical variant for a specific human disease (e.g. genetic disease), assigning a patient clinical variant to a pathogenicity category (e.g. pathogenic or benign) for the specific human disease to determine whether that patient should be followed up with additional, more invasive diagnostic testing, or treatment.