Probabilistic Modeling of Cell-Free DNA for Genetic State Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting cancer in bodily fluids face challenges due to the low amount of nucleic acids present and the contamination of tumor DNA with normal DNA, making it difficult to accurately analyze and detect specific cancer-related genetic material.
Innovation Solution
A computer technology system that uses probabilistic models to precisely diagnose genetic states from cell-free DNA, distinguishing between somatic homozygous and heterozygous deletions, and other genetic states by generating competing models based on germline single nucleotide polymorphism positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If cell-free DNA analysis is performed in bodily fluids to detect cancer, then non-invasive diagnosis is achieved, but the low amount of nucleic acids and contamination with normal DNA reduces measurement precision
Solution Approach 1:
The patent segments the analysis by dividing germline SNPs into different frequency categories (common vs. rare variants) and applying different computational models to each segment. This segmentation allows the system to handle the low signal-to-noise ratio by treating different SNP types with appropriate statistical approaches, thereby improving measurement precision while maintaining non-invasive cell-free DNA analysis
Solution Approach 2:
The patent introduces germline SNP allele frequencies as an intermediary parameter to mediate between the raw cell-free DNA data and the final cancer detection result. By using population-based allele frequency data as an intermediary reference, the system can distinguish tumor-derived DNA from normal DNA contamination, improving detection accuracy without requiring invasive procedures
2Measurement precision
If computational models analyze germline SNP positions to distinguish genetic states, then precision diagnosis of genetic material states is improved, but device complexity increases
Solution Approach 1:
The computational system is segmented into distinct modules: a data processing module that handles raw sequencing data, a probabilistic modeling module that applies statistical models to germline SNPs, and a diagnosis module that outputs genetic state predictions. This segmentation manages complexity by organizing the computational workflow into independent, manageable components while maintaining high precision in genetic state diagnosis
Solution Approach 2:
The patent changes parameters by using population-based allele frequency distributions as input parameters for the probabilistic models. By incorporating these external reference parameters, the system achieves precise genetic state diagnosis without requiring overly complex internal model structures, thus balancing precision with manageable system complexity
Data Source
AI summary
The disclosure relates to computer technology for precision diagnosis of various states of genetic material such as a gene sequenced from cell-free DNA in a sample. The state may include a somatic homozygous deletion, a somatic heterozygous deletion, a copy number variation, or other states. A computer system may generate competing probabilistic models that each output a probability that the genetic material is in a certain state. Each model may be trained on a training sample set to output a probability that the genetic material is in a respective state. In some embodiments, the computer system may use various probabilistic distributions to generate the models. For example, the computer system may use a beta-binomial distribution, a binomial distribution, a normal (also referred to as “Gaussian”) distribution, or other type of probabilistic modeling techniques.