Single Cell Variant Profiling via Bulk Sequencing Mediation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Single cell analysis techniques face challenges in generating accurate variant profiles and gene expression profiles due to sparse and noisy signals, limited DNA amplification, and dropout issues in RNA-Seq protocols, leading to uneven genome coverage and reduced sequencing reads per cell.
Innovation Solution
A system and method that combines bulk sequencing data with single cell data to validate variants and generate gene expression profiles by comparing identified variants to validation data and using a projection function to bridge the gap between single cell and bulk RNA-Seq data, enhancing variant calling and gene expression estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single cell DNA-Seq is performed to analyze heterogeneity between cells, then researchers can investigate differences between subclones and discover bio-markers, but the technique suffers from bias in amplification of limited DNA resulting in uneven coverage along the genome
Solution Approach 1:
The patent combines bulk sequencing data with single cell sequencing data to create a hybrid analysis approach. Bulk sequencing provides high coverage and uniformity, while single cell sequencing provides resolution of individual cell heterogeneity. By merging these data types, the method achieves both accurate variant calling and coverage uniformity that neither approach could achieve alone.
Solution Approach 2:
The patent uses bulk sequencing data as an intermediary reference to guide and validate variant calls from single cell data. The bulk data serves as a mediator that helps correct for amplification biases and coverage unevenness in single cell data, improving the reliability of variant detection while preserving single cell resolution.
2Measurement precision
If single cell RNA-Seq protocols sequence the 3′ ends of mRNAs to enable gene expression analysis, then researchers can profile transcriptomics at single cell resolution, but Dropouts are common and reads only cover the 3′ sites resulting in sparse and noisy data
Solution Approach 1:
The patent merges bulk RNA-Seq data with single cell RNA-Seq data to compensate for information loss in single cell measurements. Bulk data provides comprehensive gene expression information that fills in dropouts and sparse regions in single cell data, enabling more accurate gene expression estimation while maintaining single cell resolution.
Solution Approach 2:
The patent performs preliminary bulk sequencing to establish a reference gene expression profile before conducting single cell sequencing. This preliminary action provides a comprehensive baseline that can be used to impute and correct for dropouts and missing data in the subsequent single cell measurements, reducing data sparsity.
3Measurement precision
If more reads are sequenced for each cell to improve signal quality, then variant calling accuracy improves, but the cost of sequencing thousands of cells increases significantly
Solution Approach 1:
The patent merges sequencing efforts by performing bulk sequencing on pooled cells alongside or instead of deep sequencing of individual cells. This approach achieves comparable or superior variant detection accuracy at lower cost by leveraging the statistical power of pooled samples while still enabling single cell-level analysis through computational methods.
Solution Approach 2:
The patent uses bulk sequencing data as a copy or reference that can be computationally decomposed and assigned to individual cells. Instead of directly sequencing each cell deeply, the method creates a bulk copy of the genetic material, sequences it, and then uses computational algorithms to infer individual cell variants, significantly reducing sequencing costs while maintaining accuracy.
Data Source
AI summary
A system (400) configured to generate a variant profile and a gene expression profile from a single cell sample, comprising: variant validation data and gene expression comparison data; single cell DNA sequencing data comprising a plurality of verified variants; single cell RNA sequencing data comprising a gene expression profile for the sample; a processor (420) configured to: (i) validate the identified variants using the variant validation data by: comparing the identified variant to the validation data; and assigning a validated classification status to the variant if the variant corresponds to the validation data; (ii) compare the obtained gene expression data to the obtained expression comparison data; and (iii) generate, based on the comparison and using a projection function, a final gene expression profile for the single cell sample; and a user interface (440) configured to provide a report comprising the identified variants and the generated final gene expression profile.


