In this talk, Dr. Elizabeth Tseng demonstrates a throughput increase for the scIso-Seq method by concatenating single-cell molecules, increasing yield a minimum of 6-fold per SMRT Cell 8M. She explains the bioinformatics workflow for analyzing concatenated scIso-Seq data, which begins with de-concatenation, followed by tagging of UMI and barcode information that can be processed by the isoseq3 pipeline for deduplication. Reads are then aligned against the reference genome, followed by SQANTI3 for transcript classification against a reference annotation (ex: GENCODE) which produces an isoform-level sparse matrix to be analyzed with single-cell tools such as Seurat. She also shows how to apply isoform-level phasing using a tool called IsoPhase, by first calling SNPs based on the deduplicated aligned reads, then using the full-length read information to identify the haplotypes, and finally assigning each isoform-haplotype its associated read counts. Since each read can be associated with its distinct cell type, this allows quantifying isoform expressions in an allele-specific and cell type-specific manner. Finally, she shows how this can be visualized using both standard genome browsers like IGV and long read-aware tools such as SWAN.
October 28, 2021 | Presentation