Delineating genotypes and phenotypes of individual cells from long-read single cell transcriptomes

Area of Science:

• Genomics and Bioinformatics
• Single-cell Multi-omics Analysis

Background:

• Single-cell nanopore sequencing of full-length mRNAs (scNanoRNAseq) offers transformative potential for multi-omics studies.
• Existing scNanoRNAseq methods face computational complexity and reliance on short-read data for accurate analysis.
• There is a need for robust computational tools to directly analyze long-read transcriptomes for both genotype and phenotype information.

Approach:

• Development of scNanoGPS, a comprehensive toolkit for analyzing same-cell genotypes and phenotypes directly from long-read transcriptomes.
• Application of scNanoGPS to 23,587 long-read transcriptomes from tumor and cell line samples.
• Standalone deconvolution of error-prone long-reads into single cells and molecules, enabling simultaneous genotype and phenotype assessment.

Key Points:

• scNanoGPS accurately deconvoluted long-read transcriptomes, enabling direct analysis of single cells and molecules.
• Distinct combinations of isoforms (DCIs) were identified between tumor, stroma, and immune cells.
• Cell-type-specific mutations were discovered, including VEGFA in tumor cells and HLA-A in immune cells, highlighting functional roles.

Conclusions:

• scNanoGPS facilitates the direct application of single-cell long-read sequencing for comprehensive multi-omics analyses.
• The toolkit enables the discovery of cell-type-specific functional elements and mutations within complex biological samples.
• This approach enhances our understanding of cellular heterogeneity and its implications in diseases like cancer.