Duplex Sequencing Indexing for Single-Cell Low-Frequency Variant Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sequencing technologies struggle to detect low-frequency somatic mutations in a high-throughput manner with high accuracy due to high error rates, complicating the analysis of biological variants associated with diseases like cancer.
Innovation Solution
The use of Duplex Sequencing (DS) with error-correction methods to identify and quantify low-frequency genetic variants by comparing sequence reads from both strands of double-stranded nucleic acids, combined with cell-specific indexing and barcoding to ensure accurate variant detection at the single-cell level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If current sequencing technologies are used to detect somatic mutations, then high throughput sequencing can be performed, but the error rate is too high to accurately detect low-frequency variants
Solution Approach 1:
The patent segments the sequencing process into two independent strands (forward and reverse), sequencing each strand separately and then comparing results. This segmentation allows error correction by identifying discrepancies between strands, thereby improving measurement precision while maintaining high throughput capability
Solution Approach 2:
The patent introduces an intermediary error-correction step that compares sequence reads from both strands before final variant calling. This intermediary process acts as a filter to eliminate sequencing errors, improving accuracy without sacrificing the high throughput nature of the sequencing
2Speed
If standard sequencing methods are used, then sequencing can be performed quickly, but sequencing errors cannot be distinguished from true biological variants
Solution Approach 1:
The sequencing process is divided into separate forward and reverse strand sequencing operations, allowing errors to be identified and corrected through comparison. This maintains speed by parallelizing the process while improving reliability through error differentiation
Solution Approach 2:
The patent implements a feedback mechanism where sequence reads from both strands are compared and used to correct errors in the final consensus sequence. This feedback loop ensures that sequencing errors are distinguished from true variants without significantly impacting the overall sequencing speed
Data Source
AI summary
Systems, devices, kits, and methods for generating sequencing libraries for detecting low-frequency variants, and for distinguishing low-frequency biological variants from low-frequency technical variants, with duplex sequencing. Methods include introducing multiple levels of indices to cellular nucleic acid molecules of cells of a biological sample to enable differentiation of a cell from other cells of the sample, differentiation of a double-stranded cellular nucleic acid molecule from others of the cell, and differentiation of the individual strands of the double-stranded cellular nucleic acid molecule. Technical variants introduced during the workflow are identified as such due to their occurrence in only one family of duplex strands of the sequencing library, and biological variants are detected due to their occurrence in both families of duplex strands of the sequencing library. Detected biological variants can be linked with a particular cell of the biological sample in a high-throughput manner.