A New Method for Nanopore Direct RNA-Seq Pushes the Limits of Long Transcript Identification and Isoform Resolution

Direct RNA sequencing (DRS) on nanopore platforms promises something short-read methods cannot: native RNA molecules read end-to-end, with isoforms, long transcripts, and RNA modifications preserved. In practice, however, many labs find that conventional DRS workflows underdeliver on that promise—especially for transcripts longer than a few kilobases.

This article outlines a recent benchmarking study that shows preparing DRS libraries at a lower temperature (30°C) with a structure-tolerant enzyme, UltraMarathonRT® (uMRT), generates higher library yields, longer reads, better recovery of longer transcripts, and more isoforms per gene, without sacrificing read quality or compatibility with existing nanopore workflows.

Low‑temperature reverse transcription (30 °C)

The current recommended protocol for DRS utilizes Induro™ RT and requires reverse transcription at 60°C. Heat accelerates RNA hydrolysis (particularly in the presence of magnesium-containing buffers) and can reduce sample integrity, especially for long, fragile transcripts. uMRT operates efficiently at 30°C, preserving RNA integrity and base modifications during RNA library prep, avoiding the degradation seen at elevated temperatures.

RNA Stability

Improved long-transcript representation in Oxford Nanopore direct RNA sequencing with UltraMarathonRT. George Maio, Li-Tao Guo, Sara Olson, Brenton R. Graveley, Jason G. Underwood. bioRxiv 2025.12.10.693495; doi:https://doi.org/10.64898/2025.12.10.693495

Longer reads and higher CPMs

The subtle, yet powerful combination of uMRT’s low operating temperature and intrinsic helicase activity allows for long, highly structured transcripts to be copied more completely. In a head‑to‑head benchmarking using total human brain RNA, compared to the protocol with Induro, uMRT yielded longer mean/median/N50 read lengths, ~25% higher RNA–cDNA hybrid yield pre‑sequencing, and increasing mapped CPM gains in ≥5 kb, ≥10 kb, and ≥15 kb bins while maintaining the same Q-scores.

Mapped CPM Figure

Better recovery of long transcripts and isoforms

Increased mapped CPMs for longer read lengths directly translate into increased identification of transcripts ≥10 kb, ≥15 kb, and ≥20 kb. When comparing the 10 longest transcripts identified, the 10th longest transcript identified by uMRT was longer than the longest transcript identified by Induro. Isoform analysis showed uMRT identified more isoforms per gene, 24–28% more intron‑retaining transcripts, and 15–20% more novel canonical splice sites.

RNA Connect 4

RNA Connect 5

Seeing more of the native transcriptome

For researchers using direct RNA sequencing on platforms from Oxford Nanopore, these results underscore a simple point: what you recover is limited by what the RT can traverse. A low-temperature, structure-tolerant enzyme like uMRT shifts that boundary, revealing long transcripts and isoforms that were previously under-represented.