Transfer RNAs, or tRNAs, are small but essential molecules that help cells turn genetic information into proteins. Beyond this classic role, changes in tRNA levels, chemical modifications, or charging with amino acids have been linked to cancer, neurodegeneration, and other diseases. Understanding these molecules in detail has been difficult, largely because many tRNAs are extremely similar in sequence and hard to measure accurately.

Researchers at the University of Hamburg describe a new approach that tackles these challenges using Nanopore-based direct RNA sequencing. This technology can read RNA molecules directly, without converting them into DNA, making it possible to study tRNA abundance, modifications, and aminoacylation status all at once. However, until now, technical limitations have restricted how reliable and scalable these measurements could be.

The new method, called ADAM-tRNA-seq, addresses two major problems. First, the team developed an RNA-based barcoding strategy that allows multiple samples to be sequenced together and accurately separated afterward. The barcode is built into the sequencing adapter and is recognized during basecalling, improving demultiplexing accuracy and throughput.

Scheme of the RNA-based barcoding and demultiplexing used in the ADAM-tRNA-seq

(A) Library preparation using four RNA barcodes (color-coded). (B) Sequencing and demultiplexing of tRNAs. Total basecalled reads are represented with color-coded adapters at left (5′ adapter) and right (3′ adapter) ends of the reads. Detected reads without 5′ or any adapter are also depicted.

Second, the researchers introduced a hierarchy-based mapping strategy to deal with the fact that many tRNAs share nearly identical sequences. Instead of forcing each sequencing read to map to a single exact tRNA, reads can be classified at different biological levels, such as isodecoder, isoacceptor, or isotype. This reduces read loss caused by ambiguous mapping and leads to more accurate quantification of tRNA populations.

The team validated ADAM-tRNA-seq using both synthetic tRNAs and complex human tRNA samples. After systematic optimization, the approach achieved classification precision of up to 99 percent. This level of accuracy makes it practical to compare tRNA pools across different conditions, tissues, or disease states.

Overall, the work shows how careful computational and experimental design can significantly improve direct RNA sequencing of challenging molecules like tRNAs. By enabling more accurate and scalable measurements, ADAM-tRNA-seq opens the door to deeper insights into how tRNA biology contributes to health and disease.

Availability – https://github.com/danielkoester91/ADAM-tRNA-seq