A single molecule and long-read view of the diverse human transcriptome

The human transcriptome is extremely complex with over 100,000 transcripts presently described for ~20,000 protein coding genes. RNA sequencing has become a powerful and ubiquitous tool for understanding gene expression and for identification of individual splice junctions within transcripts1-7. However, short reads generated by current technologies do not allow a full-length view of transcript isoforms. Thus, a complete understanding of all spliced RNAs within a transcriptome is beyond our grasp and can only be inferred from a patchwork of short fragments. Sequencing full-length cDNA molecules in an amplification-free manner can yield an accurate understanding of isoform structures and their relative expression levels. Here, we employ Pacific Biosciences' long-read sequencing platform8 to investigate the isoform complement of a pool of diverse RNA-samples. We demonstrate that many novel transcripts and isoform structures exist within the human transcriptome and provide a more comprehensive and quantitative view of its true complexity.

人类转录组极为复杂，目前已在约20000个蛋白质编码基因中鉴定出超过100000种转录本。RNA测序（RNA sequencing）已成为解析基因表达、鉴定转录本内个体剪接接头的高效且普及的研究工具1-7。然而，当前测序技术产生的短读长片段无法提供转录本异构体的完整视图。因此，完整解析转录组内所有剪接型RNA的全貌远超当前研究能力，仅能通过碎片化的短读段片段拼凑推断得到。以无扩增方式对全长cDNA分子进行测序，可精准获取异构体结构及其相对表达水平的准确信息。本研究采用太平洋生物科学（Pacific Biosciences）公司的长读长测序平台8，对混合的多样化RNA样本的异构体组成展开探究。我们证实人类转录组中存在大量全新转录本与异构体结构，并为其真实复杂性提供了更为全面且定量的解析视角。