AMmRNA_BBC: mRNA Subcellular Localization Data for Model Training

A dataset supporting a novel multi-stage model for mRNA subcellular localization prediction. The data includes multimodal features such as nucleotide pairs, subcellular abundance-position rates, and protein-protein interaction data, evaluated on the CeFra-Seq and RNALocate datasets. The dataset was authored by mojtaba zolfi and last updated on 2026-04-16.

Use Cases

• Training classifiers for mRNA subcellular localization based on nucleotide pair features.
• Developing models to handle class imbalance in biological data using oversampling techniques.
• Extracting relationships between multimodal features like PPI data and mRNA location patterns.
• Benchmarking new bioinformatics models against the CeFra-Seq and RNALocate datasets.

Strengths

• Dataset size is 55.4 MB, indicating a contained, manageable download.
• Includes multimodal features such as nucleotide pairs and protein-protein interaction data.
• Released under a permissive CC-BY-4.0 license.

Limitations

• Row count and column definitions are unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Features were extracted from biological sequences and PPI data for model development.

Freshness

Last updated 2026-04-16 10:51:27; freshness should be verified.