How RNA Preparation Strategies Shape Direct RNA Sequencing Results


Oxford Nanopore direct RNA sequencing enables direct characterization of native RNA molecules, but upstream RNA preparation steps can strongly impact transcript representation and detection. Using the Human High-Input riboPOOL in A549 cells, we investigated how rRNA depletion, size selection, and in vitro polyadenylation shape direct RNA sequencing libraries.

Key findings at a glance

  • rRNA depletion dramatically reduces rRNA-derived reads and increases the proportion of informative transcripts.
  • Retaining short fragments preserves lncRNAs and snRNAs, whereas selection of fragments >200 nt enriches for mRNAs.
  • In vitro polyadenylation expands RNA profiling beyond conventional poly(A)+ transcripts by enabling detection of non-polyadenylated RNA species.
  • Without in vitro poly(A) tailing, lncRNA recovery is strongly reduced, indicating that many lncRNAs are not naturally polyadenylated.
  • rRNA depletion using High-input riboPOOLs increases informative transcript coverage without introducing transcript length bias.

rRNA depletion increases access to informative transcripts

In non-depleted samples, rRNA accounted for approximately 90% of mapped sequencing reads, leaving limited capacity for detection of biologically relevant transcripts. Following depletion with the Human High-Input riboPOOL, rRNA-derived reads were substantially reduced, resulting in a strong increase in the relative representation of mRNAs and non-coding RNAs.

During cleanup (Zymo RNA Clean & Concentrator kit), samples were processed either as total RNA (>17 nt) or size-selected for fragments >200 nt. Keeping total RNA maintained broad RNA diversity, including lncRNAs and snRNAs. Size selection for fragments >200 nt shifted the library composition toward longer transcripts, increasing mRNA representation while reducing shorter RNA species.

Polyadenylation reveals a hidden population of non-polyadenylated lncRNAs

A major effect was observed when comparing depleted libraries with and without in vitro polyadenylation. Without additional poly(A) tailing, lncRNA detection was strongly reduced, demonstrating that many lncRNAs are not efficiently captured by workflows relying on native poly(A) selection.

This finding highlights an important biological aspect of transcript complexity: although many sequencing strategies focus on polyadenylated RNA populations, a substantial fraction of lncRNAs appears to lack endogenous poly(A) tails. In vitro polyadenylation therefore enables direct RNA sequencing to access RNA populations that would otherwise remain largely invisible.

rRNA depletion preserves transcript length representation

Importantly, rRNA depletion using High-Input riboPOOLs did not introduce detectable transcript length bias. Mean mRNA lengths before and after depletion showed an almost perfect correlation (R = 0.996, p < 2.2 × 10⁻¹⁶), confirming that the workflow removes rRNA efficiently while maintaining transcript integrity and representation.

Conclusion

Ribodepletion using High-Input riboPOOLs improves Oxford Nanopore direct RNA sequencing by reducing rRNA background while preserving transcript characteristics. Combined with appropriate cleanup-based size selection and in vitro polyadenylation (or omission thereof), the workflow should be optimized to enhance detection of the RNA species of interest.


Want to know more? You can find more information and the full list of available High-Input riboPOOLs here. Would you like to get in contact with our product specialist or request a quote? You can reach us at info@sitools.de