bioecon Member area

Direct RNA: reading the molecule as it is

How a protein pore turns an ionic current into a native-RNA sequence, why the missing copy step changes what transcriptomics can claim, and where the chemistry is still research-grade.

Explanatory reference: fundamental scientific principles, process chemistry, and frontier mechanisms. For validated industrial metrics, suppliers, and tiered literature sources, see the linked technology dossier. How this reference is made · Report an error

Transcriptomics usually begins with a translation step: RNA is reverse-transcribed into cDNA and amplified, because sequencing platforms read DNA efficiently and read RNA poorly. Direct RNA sequencing removes the translation. The native strand is pulled through a protein nanopore one molecule at a time, each passing base perturbs an ionic current, and the current trace is decoded into sequence while the read is still in flight. Nothing is copied, so nothing can be copied wrong: the reverse-transcription bias that under- and over-represents transcripts, and the amplification that flattens relative abundance, never enter the measurement. The read is full-length by construction, which is why isoform resolution — distinguishing transcripts that share exons but differ in splicing — is a native property rather than an assembly inference. The deeper consequence is chemical. A copied sequence is an interpretation; a native strand carries its own modifications, and m6A-class base modifications perturb the ionic current in characteristic ways. Modification mapping therefore rides on the same read that yields the sequence instead of requiring a separate chemistry, and the field’s long-standing separation between “what the transcript says” and “what the transcript carries” begins to close. The corpus’s ledger record shows the chemistry operating at two altitudes. Oxford Nanopore — the UK-listed platform owner (LSE: ONT), 1,314 employees — builds the consumable base for it: more than a million flow cells a year from the Harwell Campus and Spectrum facilities, manufacture under ISO 13485, and a GridION Dx configuration marked CE-IVD and UKCA that marks the regulated edge of a research-first platform. Genoscope, the French national sequencing centre, shows the deployment altitude: nanopores reading ocean and soil metagenomes in real time, the nanoMDBG assembler turning reads into genomes as the run proceeds, and the €41.3 million ATLASea program (2024) giving the work a marine-genome mandate under CEA, France Génomique and ANR funding. The honest limits sit in three places. The chemistry is research-first: the diagnostic configuration is the exception that proves the framing, and native-RNA transcriptomics runs under research-use expectations in the US regulatory frame. Sensitivity is a physical argument — a pore reads what is in the sample, without amplification to rescue scarce transcripts — so low-input work rides on an updated cDNA-PCR kit, which re-enters some of the bias the direct route exists to remove. And the deployment record is thin: two ledger-backed organizations carry this page, which reflects verified coverage, not market size; the platform-mix detail in Genoscope’s record — Oxford Nanopore pores beside Illumina and MGI short-read instruments — is the honest picture of a complementary technology, not a replacement. The boundary against sibling pages is layer-drawn: long-read-sequencing-services owns the service market and the DNA long-read rows; spatial-transcriptomics owns where molecules sit in tissue; this page owns the chemistry of reading RNA as it exists.

Evidence and practice

Nanopore direct RNA sequencing — the technology dossier: sources with check dates, maturity, what it replaces, how to implement it, who delivers it.