MENU

tRNA Analysis: Deep Learning for Nanopore Signal Identification

Nikkei Japan
Overview
Researchers from the University of Tokyo developed a deep learning method to identify tRNA types, quantities, and modifications directly from nanopore current signals. The model achieved 98.2% accuracy in classifying all 48 types of tRNA in E. coli, providing a foundation for studying protein synthesis and disease.
In Depth

Key Findings

A research group led by Special Lecturer Hiroaki Ueda and Professor Tsutomu Suzuki from the University of Tokyo has developed a technology to integratedly analyze the type, abundance, and modification state of transfer RNA (tRNA). Using a deep learning model called tyCooNN, the team created a “nano tRNA classifier” that identifies tRNA types directly from current signals measured by a nanopore sequencer. In tests using E. coli, the model achieved a classification accuracy of 98.2% across all 48 tRNA types.

Technical Details

The developed method bypasses “basecalling”—the process of converting current signals into base sequences—and instead analyzes the raw current patterns. To address the challenge of quantification, the researchers established an absolute quantification method. By using a standard sample consisting of purified tRNAs mixed in equal molar amounts, they were able to correct for biases in reading efficiency specific to each tRNA species. This allowed for the comparison of different tRNA abundances using a common baseline.

Additionally, the team developed a technique to determine the presence or absence of specific RNA modifications at the single-molecule level. By training the model on signals from tRNAs with and without specific modifications, they demonstrated the ability to estimate the modification rate for each tRNA species.

Background & Context

tRNA molecules correspond genetic codes in messenger RNA (mRNA) to amino acids. Their abundance and modification states influence the speed and accuracy of protein synthesis, mRNA stability, and protein quality maintenance. However, because tRNAs have highly similar sequences and contain numerous chemical modifications, traditional sequencing analysis has struggled with accurate identification and quantification.

Significance & Outlook

The researchers applied this method to E. coli, capturing changes in tRNA composition associated with nutritional conditions and growth stages. This integrated analysis of tRNA quantity and modification status serves as a technical basis for investigating cellular protein synthesis and its relationship to diseases. The findings were published in Nature Communications on November 9, 2026.

Source: https://www.nikkei.com/article/DGXZRSP715197_S6A111C2000000/

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

  • Your email and selected fields are used only to deliver the newsletter.
  • We never share your information with third parties.
  • You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Takes about a minute · Unsubscribe anytime

Published by Troy-Technical, an independent site run by one engineer with a career in materials development.
About the author / Contact info@troy-technical.jp
Let's share this post !

Author of this article

TOC