Abstract
Purpose :
In recent years, a number of studies have utilised single cell transcriptomics to profile the human retina, providing important insights into the genetic signals in different retinal cell types that enable vision. However, conventional single cell RNAseq methods utilised 3’ short-read sequencing which is not suitable to identify isoform variants. Recent advancement in full-length sequencing using the PacBio isoform sequencing (Iso-Seq) technology provide an exciting opportunity to address this problem. In this study, we utilised Iso-Seq to profile human retina at single cell resolution, which allowed us to comprehensively analyse the transcriptome landscape for isoform discovery.
Methods :
Through the Lions eye banks, we collected retina samples from three healthy donors with short post-mortem time and single nuclei capture using the 10X Chromium. cDNA library was generated and processed for short-read 3’ sequencing using Novaseq following standard 10X protocol, as well as long-read sequencing using Pacbio Sequel II following the Single-Cell Iso-seq protocol. In total, we generated ~1.4B short reads and ~24.1M HiFi reads using 9 Pacbio SMRT 8M cells. The 10X and Pacbio datasets were overlaid to identify retinal cell types and assign isoforms to individual cells.
Results :
We generated a retina transcriptome dataset comprising of 25302 nuclei from major retinal cell types. Following mapping, our long-read dataset contained ~22M CCS reads and identified 34,936 unique genes, including 23,914 annotated genes and 11,022 novel genes. Structural isoform characterization using SQANTI3 classification showed that transcripts mapping to known reference accounted for ~58% of the sequencing data, while novel transcripts of known genes accounted for 37% of the dataset. We surveyed the use of alternative promoters to drive transcript variant expression, and showed that 1-8% of genes utilised multiple promoters across major retinal cell types. Also, our results enabled gene expression profiling of known and novel transcript variants for inherited retinal disease (IRD) genes, and identified differential usage of exon splicing in different retinal cell types.
Conclusions :
Here we generated a human retina transcriptome dataset at single cell resolution with full-length sequencing. Our results highlighted the potential of Iso-seq to profile transcript isoforms and study splicing in different human retina cell types.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.