Tumor Sample Analysis Using Segmented Alignment and Variable Depth Bait Sets

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for analyzing tumor nucleic acids are inefficient in handling diverse genetic events across a large number of genes, leading to suboptimal speed, sensitivity, and specificity in mutation detection.

Innovation Solution

The method integrates multiple, individually tuned alignment and mutation calling methods tailored to specific genes, tumor types, and variant characteristics, using bait sets with varying sequencing depths to enrich and analyze subgenomic intervals, optimizing the alignment and mutation detection process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single uniform alignment method is used for all genes, then the process is simple to implement, but the sensitivity and specificity of mutation detection is suboptimal

Engineering Contradiction:
Improvemutation detection sensitivity and specificityVSAvoidalignment method complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the alignment process into multiple methods tailored to different gene categories. Instead of using a single uniform alignment method for all genes, the system segments genes into groups (e.g., highly variable genes, genes with repetitive elements, standard genes) and applies specific alignment algorithms optimized for each segment's characteristics, thereby improving mutation detection precision without overwhelming complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies the principle of local quality by customizing alignment parameters and methods according to the specific characteristics of different gene regions. Each gene or gene group receives locally optimized alignment settings (e.g., different mismatch penalties, gap costs, or algorithm choices) based on its sequence properties, ensuring that each region is analyzed with the most appropriate method for its specific needs

Inventive Principle:
Principle #3Local quality

2Measurement precision

If bait sets with uniform sequencing depth are used, then the workflow is simpler, but the detection sensitivity for diverse genetic events is reduced

Engineering Contradiction:
Improvedetection sensitivity for diverse genetic eventsVSAvoidbait set design complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamic bait set design where the sequencing depth assigned to each bait set is adjusted based on the specific genetic events being targeted. Different bait sets are configured with varying depths (e.g., higher depth for detecting rare mutations, lower depth for common variants) according to the expected frequency and clinical significance of the genetic events, allowing the system to adapt its resources to match detection needs

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of sequencing depth across different bait sets rather than maintaining a uniform depth. By varying this critical parameter based on the target gene's characteristics and the type of genetic events being sought, the system optimizes detection sensitivity for diverse events while managing overall sequencing complexity through reasoned parameter differentiation

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If multiple individually tuned alignment methods are used for different genes, then the mutation detection accuracy is improved, but the computational time and processing complexity increases

Engineering Contradiction:
Improvemutation detection accuracyVSAvoidcomputational processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the genome into different regions or gene categories, each processed by an optimized alignment method. This segmentation allows parallel processing of different gene sets, reducing overall computational time compared to sequentially processing all genes with a single method, while maintaining high accuracy through method-specific optimization for each segment

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent develops a multi-functional alignment framework that can handle multiple types of genetic variations and gene characteristics within a unified computational architecture. This universal system incorporates multiple alignment strategies but manages them through a single coordinated platform, reducing overhead and processing time compared to running separate independent analysis pipelines for each gene type

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

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances the speed, sensitivity, and specificity of mutation detection in tumor samples by optimizing alignment and mutation calling methods for diverse genetic events across a large number of genes, ensuring high sensitivity and specificity for clinical applications.

Implementation Method 1

each bait set is a plurality of nucleic acid molecules which can hybridize to and thereby capture a target nucleic acid

Methodology Applied
Scientific EffectHybridization: Chemical Bonding

Data Source

PatentEP3564395B1Optimization of multigene analysis of tumor samples
Publication Date: 2025.06.25 FOUNDATION MEDICINE INC
  • EP3564395B1 patent drawingFigure 1A
  • EP3564395B1 patent drawingFigure 1B
  • EP3564395B1 patent drawingFigure 1C

AI summary

A method of analyzing a tumor sample includes acquiring a library comprising a plurality of tumor members from the sample, contacting the library with a bait set to isolate selected members, acquiring a read for a sub-genomic interval from a selected member, aligning said read and assigning a nucleotide value (e.g., calling a mutation) from said read for a preselected nucleotide position