Tumor Variant Amplitude Scoring for Cancer Drug Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack the ability to effectively prioritize which gene variants should be targeted by drugs in cancer treatment, as existing prediction tools fail to provide evidence on the usefulness and extent of different types of data for distinguishing disease-driving variants from neutral variants.
Innovation Solution
A method using a computer system to analyze genomic databases, calculate a Tumor Variant Amplitude (TVA) based on the ratio of observed to expected gene variant occurrences, and filter data to assess the biological effect of gene variants, enabling better forecasting of treatment responses and prioritization of variants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple types of data are used to distinguish disease-drivers from neutral variants, then the identification accuracy improves, but the complexity of analysis increases
Solution Approach 1:
The patent segments the complex analysis into distinct functional modules: data collection module, data processing module, TVA calculation module, and prioritization module. Each module handles specific aspects of the analysis, making the overall system more manageable and interpretable while maintaining high identification accuracy through multi-data integration.
Solution Approach 2:
The patent introduces Tumor Variant Amplitude (TVA) as an intermediary metric that synthesizes information from multiple data types (observed occurrences, expected occurrences, mutation rates). This intermediary simplifies the complex multi-data analysis into a single quantitative measure that can be directly used for prioritization, reducing analysis complexity while preserving identification accuracy.
2Reliability
If all identified variants are targeted by drugs, then comprehensive coverage is achieved, but side effects and drug availability issues arise
Solution Approach 1:
The patent changes the parameter of variant prioritization from binary (targeted/not targeted) to continuous (TVA scores). By calculating TVA for each variant and setting prioritization thresholds, the system identifies a subset of high-impact variants for drug targeting. This parameter transformation enables selective targeting of the most critical variants, maintaining treatment effectiveness while reducing exposure to unnecessary drugs and minimizing side effects.
Data Source
AI summary
The techniques described herein disclose a method or a system for analyzing genomic data, calculating a predictor and making a quantitative assessment of a biological effect based on the predictor. A biological effect such as the pathogenicity of a cancer a risk that a subject may develop a particular cancer may be determined based on the predictor. The predictor may comprise the observed number of occurrences of a gene variant divided by the expected number of occurrences of the gene variant. The prediction of a drug treatment may comprise prioritization of gene variants according to a selective variant effect and determining which drug treatment to prioritize. The predictions may further comprise using genomic coordinates for each gene variant and nucleotide alterations from various databases, but filtering out duplicate samples from the same subject.