Compound Promiscuity Data Sets for Multi-Target Activity Predictions

Three data sets contain compounds with multi- or single-target activity assembled from the PubChem BioAssay database. They were deposited by Christian Feldmann of the University of Bonn for use in promiscuity predictions. The data structure is described in an accompanying readme file and a forthcoming data note.

Use Cases

• Train machine learning models to predict compound promiscuity based on bioassay activity data.
• Analyze patterns of multi-target activity among chemical compounds.
• Benchmark computational methods for identifying promiscuous compounds.

Strengths

• Data is sourced from the authoritative PubChem BioAssay database.
• Includes three distinct data sets for comparative analysis.

Limitations

• Row count and file size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

PubChem BioAssay database

Collection Method

Assembled from the PubChem BioAssay database.