DNABERT_vectors: Plasmid and Chromosome Feature Embeddings

DNABERT embeddings calculated from plasmids and chromosomes. Maho Tokuda created this 2.1 MB dataset for a RandomForest model predicting plasmid destinations. The dataset was last updated in June 2026.

Use Cases

• Train plasmid destination prediction models based on DNABERT-derived feature vectors.
• Benchmark machine learning algorithms for genomic sequence classification based on the described embeddings.
• Analyze feature representations of plasmids versus chromosomes based on the dataset's source.
• Develop transfer learning models for other genomic tasks using the pre-computed DNABERT vectors.

Strengths

• Dataset size is precisely 2.1 MB, indicating a manageable download.
• License is explicitly CC-BY-4.0, permitting broad reuse.
• Last update date is June 2026, suggesting recent availability.

Limitations

• Row count is unknown, which may limit 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

Calculated from plasmids and chromosomes using DNABERT.

Freshness

Last updated 2026-06-01 01:54:55; freshness should be verified.