Summary

Oxford Nanopore direct RNA sequencing preserves native RNA molecules by sequencing them directly — no reverse transcription, no PCR amplification. This makes it the method of choice for detecting RNA base modifications, measuring poly(A) tail lengths, and characterizing full-length transcript isoforms. But the lack of amplification also means the library quality depends entirely on the input RNA. Degraded RNA, insufficient poly(A)+ content, contaminants, or low input amounts all translate directly into reduced throughput, shorter reads, and biased data. This guide covers the sample QC requirements that determine whether a direct RNA sequencing experiment succeeds. If you are preparing samples for direct RNA sequencing, explore the Nanopore Direct RNA Sequencing service for detailed sample submission guidelines.

Figure 1: Sample QC flow for ONT direct RNA sequencing — from RNA extraction through integrity assessment, poly(A) selection, and purity verification before library preparation.

Key Takeaways

• Direct RNA sequencing requires intact RNA (RIN ≥ 8) because there is no amplification step to rescue degraded molecules
• Poly(A)+ selection is required — RNA without poly(A) tails cannot be sequenced directly
• Minimum input is ~300 ng poly(A)+ RNA or ~1 µg total RNA; less input reduces throughput proportionally
• Contaminants (phenol, salts, ethanol, EDTA) block nanopores and cause run failure more often than degraded RNA
• Degraded RNA produces 3′-biased reads and underestimates poly(A) tail lengths

Why Direct RNA Sequencing Has Different QC Requirements

Unlike cDNA-based sequencing methods, direct RNA sequencing has no reverse transcription or PCR amplification step. The native RNA molecule itself is threaded through the nanopore, and the current modulation is measured directly. This has two consequences for sample QC.

First, every RNA molecule in the library must be capable of interacting with the nanopore. Adapter ligation depends on the presence of a poly(A) tail at the 3′ end, and the RNA backbone must be intact enough to be processed through the pore. Damaged or truncated molecules that could produce usable cDNA after reverse transcription will produce low-quality or zero signal in direct RNA sequencing.

Second, contaminants in the sample cannot be diluted out by amplification. A PCR step can tolerate some level of carryover impurity because the polymerase selectively amplifies the template. Direct RNA sequencing has no such tolerance — contaminants that block nanopores or interfere with the motor protein reduce the number of active pores and lower throughput for the entire run.

These differences make sample QC for direct RNA sequencing more stringent than for cDNA-based RNA sequencing, and the QC gates must be checked before committing precious samples to library preparation.

RNA Integrity: The Most Important Parameter

RNA integrity — measured as RIN (RNA Integrity Number), RQN (RNA Quality Number), or TIN (Transcript Integrity Number) — is the single most important predictor of direct RNA sequencing success.

Why integrity matters for direct RNA. The direct RNA sequencing workflow ligates adapters to the 3′ end of RNA molecules and then threads the 3′ end into the nanopore. If the RNA is fragmented, only the fragment containing the original 3′ end will have a poly(A) tail and can be captured. The 5′ fragments are lost. This creates an unavoidable 3′ bias in the data — and the more degraded the RNA, the stronger the bias.

The practical thresholds are:

For studies focused on RNA modification detection or poly(A) tail length analysis, the integrity requirement is more stringent because both analyses depend on reads spanning the full length of the transcript. A RIN of 8 or higher is the minimum for modification detection at single-base resolution, and a TIN above 70 is recommended for poly(A) tail measurements.

Assessing integrity. Use microfluidic electrophoresis — Agilent Bioanalyzer, TapeStation, or Qsep — to generate an electropherogram and RIN/RQN score. The electropherogram itself should be inspected: a clean trace with distinct 18S and 28S ribosomal RNA peaks (for total RNA) indicates good integrity, while a shift toward shorter fragment sizes and a loss of ribosomal peaks signals degradation.

Figure 2: Bioanalyzer electropherogram comparison of intact (RIN 9.2) versus degraded (RIN 4.8) total RNA, showing the loss of ribosomal peaks and the shift toward shorter fragments in degraded samples.

Poly(A) Status: Required, Not Optional

Direct RNA sequencing depends on the poly(A) tail for adapter ligation. The sequencing adapter includes a poly(T) stretch that hybridizes to the poly(A) tail, and this hybridization is what positions the RNA molecule for entry into the nanopore.

Poly(A)+ RNA is required. RNA samples that lack poly(A) tails — most prokaryotic RNA, non-polyadenylated eukaryotic transcripts (e.g., histone mRNAs), and RNA that has been deadenylated during extraction or storage — cannot be sequenced directly without modification. If the sample is known to lack poly(A) tails, in vitro polyadenylation using E. coli poly(A) polymerase can be performed before library preparation, but this adds a step and may introduce bias.

Poly(A) tail length considerations. The natural poly(A) tail length in eukaryotic mRNA varies from approximately 50 to 250 adenosines. ONT’s direct RNA kit (SQK-RNA004) is designed to work with this range. Very short poly(A) tails (below 20 nt) may hybridize inefficiently to the adapter, reducing capture efficiency. Very long tails (above 300 nt) can cause the motor protein to stall at the poly(A) tract, producing reads that consist mostly of A homopolymers.

For most eukaryotic total RNA or poly(A)-selected RNA samples, the native poly(A) tail distribution is adequate. If the sample has been subjected to RNA treatments or chemical modifications that could alter poly(A) tails, a poly(A) tail length assay before library preparation is a useful risk-reduction step.

Input Amount: Minimums and Practical Limits

The input amount for direct RNA sequencing is higher than for cDNA-based methods because there is no amplification. Every molecule in the library is a single molecule that must find and interact with a nanopore.

The relationship between input amount and throughput is approximately linear within the practical range: doubling the input roughly doubles the number of reads, up to the saturation point of the flow cell. Below 200 ng of poly(A)+ RNA, throughput drops sharply, and the number of reads may be too low for differential expression or modification detection.

Sample volume is also a constraint. The library preparation requires the RNA to be in a small volume — typically 8–10 µL for the adapter ligation step. If the RNA is too dilute, it must be concentrated, which carries its own risks of degradation and loss. Submitting RNA at a concentration of at least 50 ng/µL for poly(A)+ RNA (or 100 ng/µL for total RNA) is a practical target.

Purity: What Blocks Nanopores

Purity requirements for direct RNA sequencing are more stringent than for most other RNA sequencing methods because contaminants cannot be removed after library preparation and directly interfere with nanopore function.

Critical contaminants to avoid:

Purity targets:

• A260/280: ~2.0 (acceptable range: 1.9–2.1)
• A260/230: 2.0–2.2 (values below 1.8 indicate contaminant carryover)
• Quantification: Use fluorescence-based methods (Qubit RNA HS Assay Kit) rather than Nanodrop alone, because degraded RNA and contaminants both inflate absorbance-based measurements

If the A260/230 ratio is below 1.8, an additional cleanup step (RNA binding bead cleanup or ethanol precipitation with a 70% ethanol wash) should be performed before library preparation. Submitting samples with a known low A260/230 ratio is the most common preventable cause of direct RNA sequencing failure.

Sample Types and Compatibility

Not all RNA sample types are equally suitable for direct RNA sequencing. The following guidance helps match sample types to realistic expectations.

High-quality total RNA from cells or tissue. This is the ideal starting material. A RIN of 8 or higher and sufficient input mass are both achievable. Poly(A) selection during library preparation isolates the mRNA fraction. For mammalian cell pellets or flash-frozen tissue, this is the sample type with the highest success rate.

Poly(A)-selected mRNA. Submitting pre-selected poly(A)+ RNA saves the poly(A) selection step during library preparation and allows a more precise input amount. The trade-off is that poly(A) selection itself can introduce degradation if not performed carefully, so the RIN should be rechecked after selection.

FFPE RNA. FFPE-derived RNA is typically degraded (DV200 below 50%, RIN below 5) and is not suitable for standard direct RNA sequencing. The fragmented RNA produces strong 3′ bias, and the chemical modifications from formalin fixation can interfere with nanopore current readings. FFPE RNA should be directed to cDNA-based nanopore sequencing or an alternative RNA sequencing method.

Viral RNA. Direct RNA sequencing of viral genomes depends on whether the viral RNA has a poly(A) tail. Positive-sense RNA viruses (e.g., SARS-CoV-2, Zika) with polyadenylated genomes can be sequenced directly. Viral RNA without poly(A) tails requires in vitro polyadenylation. The low abundance of viral RNA in clinical samples may also require enrichment strategies such as antisense oligo pull-down.

Total RNA from prokaryotes. Bacterial and archaeal RNA lacks poly(A) tails and cannot be sequenced directly without in vitro polyadenylation. Even with polyadenylation, the throughput and data quality are lower than for eukaryotic poly(A)+ RNA.

QC Checkpoints Before Library Preparation

A systematic QC workflow before committing to library preparation saves time and flow cell cost:

1. Quantify with a fluorescence-based method (Qubit RNA HS). Record concentration in ng/µL.
2. Check purity by Nanodrop. Record A260/280 and A260/230. If A260/230 < 1.8, perform additional cleanup.
3. Assess integrity by microfluidic electrophoresis. Record RIN/RQN and inspect the electropherogram for degradation smearing.
4. Check volume against the library preparation requirement. The RNA must be in ≤10 µL at the required concentration. If too dilute, concentrate by ethanol precipitation or speed-vac (with care to avoid degradation).
5. Confirm poly(A) status if the sample type is unusual or if the RNA has been chemically treated.

Samples that pass all five checks have a high probability of producing a successful direct RNA sequencing library. Samples that fail one or more checks should be flagged for protocol adjustment or method change before proceeding.

Figure 3: Five-stage QC decision flowchart for direct RNA sequencing samples — quantify, check purity, assess integrity, check volume, confirm poly(A) status.

Frequently Asked Questions

1) What is the minimum RIN score for direct RNA sequencing?

The minimum recommended RIN is 7, and RIN 8 or higher is strongly preferred. Samples with RIN below 6 should not be used for direct RNA sequencing; the data will be dominated by 3′ bias and will not support reliable modification detection or isoform analysis. For such samples, cDNA-based nanopore sequencing is a more appropriate alternative.

2) Can I sequence bacterial RNA with direct RNA sequencing?

Bacterial RNA lacks poly(A) tails and cannot be sequenced directly with the standard ONT direct RNA kit. In vitro polyadenylation can add poly(A) tails to bacterial RNA, making library preparation possible, but the throughput and data quality are typically lower than for eukaryotic poly(A)+ RNA.

3) How much does throughput change with input amount?

Throughput is approximately linear with input within the practical range of 200 ng to 1 µg of poly(A)+ RNA. Below 200 ng, throughput drops disproportionately as the molar ratio of adapter to RNA becomes unfavorable and library loss during cleanup has a larger relative effect.

4) What should I do if my A260/230 ratio is below 1.8?

Perform an additional RNA cleanup step. RNA binding bead cleanup (Ampure RNAClean or equivalent) or a lithium chloride precipitation with a 70% ethanol wash typically resolves low A260/230 ratios. The cleaned RNA should be re-quantified and rechecked before proceeding.

5) Does direct RNA sequencing work with FFPE RNA?

Standard direct RNA sequencing is not recommended for FFPE RNA. The degradation level (typically DV200 below 50%) produces strong 3′ bias and low throughput. If FFPE RNA must be used, cDNA-based nanopore sequencing or short-read RNA-seq are more reliable options.

Related CD Genomics Services

• Nanopore Direct RNA Sequencing — Direct RNA sequencing with native modification detection
• RNA Modification Service — Detection and analysis of RNA base modifications

References

1. Garalde DR, Snell EA, Jachimowicz D, et al. Highly parallel direct RNA sequencing on an array of nanopores. Nature Methods. 2018;15:201-206. doi:10.1038/nmeth.4577
2. Workman RE, Tang AD, Tang PS, et al. Nanopore native RNA sequencing of a human poly(A) transcriptome. Nature Methods. 2019;16:1297-1305. doi:10.1038/s41592-019-0617-2
3. Parker MT, Knop K, Sherwood AV, et al. Nanopore direct RNA sequencing maps the complexity of Arabidopsis mRNA processing and m⁺A modification. eLife. 2020;9:e49658. doi:10.7554/eLife.49658
4. Oxford Nanopore Technologies. Direct RNA sequencing: SQK-RNA004 kit documentation. Accessed June 2026.

Services mentioned in this article are provided for research use only and are not intended for clinical diagnosis, treatment, or personal health assessment.