Simultaneous profiling of full-length RNA transcripts and chromatin accessibility within single cells of human retinal organoids

Abstract

Single-cell multi-omics sequencing can integrate transcriptome and epigenome to analyze the complex mechanisms underlying neuron development and regeneration, but most current methods are based on second-generation short-read sequencing, which has low efficiency in detecting RNA structural heterogeneity. Long-length sequencing can analyze RNA structures, but the throughput and the number of transcripts detected at the single-cell level are very low, and single-cell level epigenome profiling has not been accomplished either. Therefore, there is currently a lack of an effective method that can integrate RNA splicing and epigenetic modification to analyze the molecular mechanism of neural development. This study developed a single-cell multi-omics assay based on short-read sequencing for the simultaneous detection of single-cell full-length RNA isoforms and DNA accessibility. The accuracy of its resolution in RNA transcript structure can reach 94.5%, and the sensitivity of detecting single-cell gene expression is twice that of third-generation sequencing. And it can detect over 10,000 single nuclei at one run, enabling the effective integrated analysis of single-cell RNA isoforms and DNA accessibility at high throughput. We used this method to construct a multidimensional cell atlas of human retinal organoids, and found that gene expression and differential choices of isoforms of multiple fate-determining factors were significantly associated with chromatin accessibility. This method provides a new technical method for dissecting the multidimensional molecular mechanism of fate determination in neural cell development and regeneration.

Advances in single-cell long-read sequencing technologies

“…It holds the potential to be incorporated into single-cell sequencing workflows, but it remains to be adopted. An innovation called scRCAT-seq (single-cell RNA cap and tail sequencing) and its improvement scRCAT-seq2 were developed to capture variation in transcripts at a single-cell level using short-read sequencing but were limited to the TSS and TTS sequencing ( 25 , 26 ). Another technology, called VASA-seq (vast transcriptome analysis of single cells by dA-tailing), alleviates the issue of bias and single-ended sequencing as observed in Smart-seq and Illumina, respectively, by polyadenylating all fragments to be sequenced ( 27 ).…”