We have been trying bacterial transcriptomics using Nanopore sequencing. I am particularly interested in direct RNA sequencing. Tonight I watched the London Calling 2023 session by Libby Snell, Principal Scientists/Applications at Oxford Nanopore Technologies, and Martin Smith from Sainte-Justine University Hospital Research Center in Canada. They presented a session entitled “Direct RNA sequencing update (SQK-RNA004): initial impressions of RNA004 and its application for functional profiling of cancer-associated long non-coding RNAs.” Snell noted that Oxford Nanopore Technologies (ONT) offers the only platform currently for direct RNA sequencing. There is no reverse transcription bias and includes an RNA specific motor protein (3′ down). Modified bases can be distinguished and can estimate poly(A) tail length. First released in 2017, the library prep takes ~120 min. While there is a cDNA synthesis step, the cDNA does not go into the pore, and the RNA is directly read. The sample prep is similar to others with the new RNA004 release and can use 100-200 enriched RNA or 1 ug of total RNA. There is also a new RNA-compatible pore called “RP4” that would offer RNA flow cells! The new motor is faster: 125 bps. With higher output because of the faster motor and neural network analysis. The increased throughput and robustness of the assay promise improved transcriptomic analyses. Snell explained that they are working on modified base detection (for RNA) and software improvements, as well as new RNA kit barcodes. The direct RNA kit has been available for six years, and the previous version (RNA002) has been significantly improved with the new RNA004 kit. The kit is still in beta testing and early access, as of the recording.
Smith spoke about testing the new RP4 pores. They obtained good pore activity and occupancy. The new pore activity and read length seem promising! Smith shared data from initial runs and then from a functional CRISPRi screen to identify growth-inhibiting lncRNAs. Washed flow cells produced useful data. Interestingly, in vitro transcribed RNA had lower error rates, possibly due to modified bases in native RNA! They noted that poly(A) enriched libraries produced slightly smaller transcripts, but after applying batch effect cleaning of the datasets produced similar trends. They sequenced 20 RNA004 PromethION runs and compared runs. To analyze the data, they did de novo meta-transcriptome assembly with Stringtie2 and identified new isoforms Nalm6 cell line. Also, Smith highlighted how the characterization of isoforms with the new kit is much cleaner. RP4 flow cells could be very useful for bacterial transcriptomics!
