Friend Or Foe: Synthetic Metabolic Data for 10,000 Bacterial Pairs

Powidla created a synthetic collection of environmental tabular datasets for machine learning. The data was generated by solving metabolic models for 10,000 pairs of bacteria. It was last updated on June 10, 2026.

Use Cases

• Predicting bacterial interaction types (e.g., Friend or Foe) based on synthetic metabolic features
• Benchmarking machine learning methods on synthetic biological data
• Training models for metabolic network analysis using generated bacterial pair data

Strengths

• Contains data for 10,000 pairs of bacteria.
• Designed for 5 different machine learning methods.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

huggingface user powidla

Collection Method

Synthetically generated by solving metabolic models.

Freshness

Last updated 2026-06-10 19:48:00