K-mer Canonical: Plasmid and Chromosome Compositions for ML Prediction

31.1 MB of k-mer compositions calculated from plasmids and chromosomes. Maho Tokuda created this dataset for a Random Forest model to predict plasmid destinations, as described in Tokuda et al., 2026. The data is available in CSV format under a CC-BY-4.0 license.

Use Cases

• Train Random Forest classifiers for plasmid destination prediction based on k-mer compositions.
• Compare k-mer frequency distributions between plasmids and chromosomes.
• Benchmark feature extraction methods for genomic sequence classification.
• Develop novel machine learning models for plasmid-host interaction studies.

Strengths

• Dataset size is 31.1 MB, indicating a substantial collection of k-mer features.
• The dataset is specifically designed for a Random Forest model, suggesting a clear application context.
• The dataset is openly available under a CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

K-mer compositions were calculated from plasmids and chromosomes.

Freshness

Last updated 2026-06-01 01:56:33; freshness should be verified.