Transcriptome profiling is one of the most frequently used technologies and key to interpreting the function of the genome in human diseases. However, quantification of transcript expression with short-read RNA-Seq remains challenging as different transcripts from the same gene are often highly similar. Nanopore RNA-Seq reduces the complexity of transcriptome profiling with ultra-long reads possibly covering the full length of the isoforms. The technology has a high sequencing error rate and often generates shorter, fragmented reads due to RNA degradation. While the increased read length improves the ability to detect novel transcripts, the high error rate observed for Nanopore RNA Sequencing data can potentially lead to wrongly assigned reads and inflated estimates of the number of novel transcripts. In order to facilitate transcript discovery and quantification from noisy long reads, we developed bambu, a long read isoform discovery and quantification method. bambu uses reference annotations to correct read alignments, assigns reads to annotated and new transcripts, and obtains high quality quantification estimates across all samples of interest. We benchmark our method using cancer cell line data with spike-in controls, showing that using reference annotations improves transcript discovery, which further improves transcript quantification.
Postdoctoral Fellow, Genome Institute of Singapore
Ying Chen is a Postdoctoral Fellow in the Lab of Computational Transcriptomics at the Genome Institute of Singapore. She holds a PhD in Public Health from the Saw Swee Hock School of Public Health at the National University of Singapore. Her research interests include biostatistics, data analytics, statistical genomics and cancer research. She is currently working on projects using Oxford Nanopore Sequencing technology, where she and her collaborators developed a reference-guided transcript discovery and quantification method, Bambu, specifically for long read RNA-Seq. Ying Chen is also working on the Singapore Nanopore Expression Project (SG-NEx), to provide a systematic comparison of different Nanopore RNA-Sequencing protocols.