NP-mFinder: Nanopore DRS Data for m6A Site Prediction

Yuening Yang published a dataset on 2026-04-21 for a computational tool called Nanopore-m6A-Finder (NP-mFinder). The tool predicts N6-methyladenosine (m6A) RNA modification sites and estimates methylation levels from Nanopore Direct RNA Sequencing (DRS) data. It was trained on in vitro-transcribed RNA DRS data and benchmarked against methods like meRIP-seq, miCLIP, EpiNano, mAFiA, and GLORI v2.0.

Use Cases

• Benchmarking m6A site callers based on performance metrics (AUC ≈0.90, F1-score >0.80) reported in the description.
• Training machine learning models for RNA modification detection using features from Nanopore DRS signals.
• Investigating m6A modifications in specific sequence contexts, such as the 'AAAAA' motif in mRNA exonic regions.
• Exploring the potential presence of m6A in poly(A) tail regions of human transcripts.

Strengths

• The associated tool achieved an area under the curve (AUC) of approximately 0.90 on held-out test data.
• Reported accuracy, precision, recall, and F1-score were all greater than 0.80.
• Demonstrated 93% precision in detecting m6A within the 'AAAAA' sequence context when compared to GLORI v2.0 annotation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is small (2.6 MB), suggesting it may contain model outputs or summary data rather than raw sequencing reads.

Provenance

Source

figshare, author Yuening Yang.

Collection Method

Data likely generated by the NP-mFinder computational framework applied to Nanopore DRS data.

Freshness

Last updated 2026-04-21 22:01:12; freshness should be verified.