HRD Detection Using Copy-Number Segments Across Chromosome Arms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting homologous recombination deficiency (HRD) are inaccurate and computationally complex due to rigid criteria for identifying large-scale state transitions (LST) and require multiple genomic factors, failing to capture genomic instability accurately and efficiently.
Innovation Solution
A method involving the identification of a first subset of genome segments with a common copy number and a second subset with varying copy numbers and lengths within a predetermined range, using a statistical model to determine HRD status based on the proportion of these segments to chromosome arms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rigid criteria for identifying large-scale state transitions (LST) are used, then measurement precision of genomic instability is improved, but device complexity and computational complexity increase
Solution Approach 1:
The genome is divided into segments along chromosome arms, and copy number profiles are segmented into distinct regions. This segmentation allows complex genomic data to be processed in manageable units, reducing computational complexity while maintaining detection precision through systematic analysis of segment characteristics
Solution Approach 2:
The patent extracts specific features from complex genomic data, including segment lengths, copy number values, and chromosomal positions. By extracting only the relevant features needed for LST identification, the method reduces computational burden while preserving the ability to accurately detect genomic instability
2Measurement precision
If multiple genomic factors are analyzed, then measurement precision of HRD detection is improved, but device complexity increases
Solution Approach 1:
The patent combines multiple genomic factors (segment lengths, copy number variations, chromosomal positions) into a unified analytical framework for detecting large-scale state transitions. By merging these factors into a cohesive LST detection algorithm, the method achieves high HRD detection accuracy without proportionally increasing complexity
Solution Approach 2:
The copy number profile analysis method serves multiple functions: it identifies LSTs, detects HRD status, and provides insights into genomic instability patterns. This multi-functionality allows a single computational approach to address multiple detection needs, reducing overall system complexity while maintaining high measurement precision
3Measurement precision
If existing LST identification methods are used, then measurement precision is improved, but productivity and computational efficiency decrease
Solution Approach 1:
The patent performs preliminary segmentation of copy number profiles and pre-identification of potential LST regions based on segment characteristics. By preparing data in advance with predefined segment criteria, the method reduces the computational workload during final LST identification, improving both precision and computational efficiency
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Techniques for determining whether a sample obtained from a subject includes cells having homologous recombination deficiency (HRD). The techniques include: obtaining data about segments of the subject's genome; identifying a first subset of the segments, the first subset including segments associated with at least one chromosome arm of the genome and having a common copy number; identifying a second subset of the segments, each of the segments of the second subset having (i) a respective copy number different from the common copy number and (ii) a respective length that satisfies a predetermined length criterion; determining a proportion of a number of segments in the second subset to a number of chromosome arms of the at least one chromosome arm; and determining, based on the determined proportion, whether the biological sample includes cells having HRD.