Basis Sets for Computing Random Neighbor Score of Protein Embeddings

A 13.3 GB collection of basis sets for computing the Random Neighbor Score of protein embeddings, authored by R Prabakaran and last updated on June 1, 2026. The dataset is shared on figshare under a CC-BY-4.0 license. Its specific row count and column structure are not detailed in the provided metadata.

Use Cases

• Evaluating the quality and statistical properties of protein embedding models based on the Random Neighbor Score metric.
• Benchmarking different protein representation learning algorithms using the provided basis sets.
• Developing novel scoring or analysis methods for protein sequence or structure embeddings.

Strengths

• Dataset size is 13.3 GB, indicating a substantial collection of basis sets.
• Shared under a permissive CC-BY-4.0 license, facilitating reuse and distribution.
• Has a specific, recent update timestamp of 2026-06-01 17:46:47.

Limitations

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

Provenance

Source

figshare, author R Prabakaran

Freshness

Last updated 2026-06-01 17:46:47; freshness should be verified.