Genome biologyMar 29, 2025

DEMINERS improves speed and accuracy of direct RNA sequencing for clinical gene and RNA analysis

Updated Mar 7, 2026

Abstract

DEMINERS can accurately demultiplex up to 24 samples while reducing RNA input and runtime.

Nanopore typically faces challenges such as low basecalling accuracy and high input requirements.
The DEMINERS toolkit combines RNA , a Random Forest-based barcode classifier, and an optimized convolutional neural network for improved accuracy.
This toolkit is associated with applications in clinical metagenomics and cancer transcriptomics.
DEMINERS may uncover microbial diversity in COVID-19 and explore roles of RNA modifications in conditions like malaria and glioma.
It provides a high-throughput solution for precise analysis of transcripts and RNA modifications.

Simplified

Nanopore (DRS) is a powerful tool for RNA biology but suffers from low basecalling accuracy, low throughput, and high input requirements. We present DEMINERS, a novel DRS toolkit combining an RNA workflow, a Random Forest-based barcode classifier, and an optimized convolutional neural network basecaller with species-specific training. DEMINERS enables accurate demultiplexing of up to 24 samples, reducing RNA input and runtime. Applications include clinical metagenomics, cancer transcriptomics, and parallel transcriptomic comparisons, uncovering microbial diversity in COVID-19 and mA's role in malaria and glioma. DEMINERS offers a robust, high-throughput solution for precise transcript and RNA modification analysis. 6

Key numbers

Demultiplexing Capacity

Number of RNA samples that can be processed simultaneously.

95.82% to 97.83%

Classification Accuracy Range

Accuracy of DEMINERS in classifying RNA samples compared to existing methods.

90%

Basecalling Accuracy

Basecalling accuracy achieved for various species using DEMINERS.

Full Text

We can’t show the full text here under this license.

View Full Text ↗

Funding

Competing interests

Declarations. Ethics approval and consent to participate: Tumour sample collection and the study design were approved by the Biomedical Research Ethics Committee of West China Hospital (Approval number: 2020.837). The research on COVID-19 specimens was approved by the Biomedical Research Ethics Committee of West China Hospital (Approval number 2020.100, 2020.193 and 2020.267). The swabs were obtained for routine diagnostic purposes, and the remaining RNA samples were provided for research. Written consents were obtained from all patients. Consent for publication: Not applicable. Competing interests: Sichuan University has filed patent applications for the methods described herein, with L.C., J.-wL., J.G., J.S., C.T., L.L., and C.C. listed as inventors. The remaining authors declare no competing interests.

PubMed

Abstract

Key numbers

Full Text

Funding

Competing interests

Related papers

What Lands in Your Inbox Each Week: