MetalCytoToxDB: 26,500 Cytotoxicity Values for Metal Complexes

MetalCytoToxDB is a manually curated database of 26,500 IC50 values for 7,050 metal complexes, tested against 754 cell lines. The data was compiled from 1,921 articles by Lev Krasnov and last updated in April 2026. It supports the development of machine learning models to predict the anticancer activity of transition-metal complexes.

Use Cases

• Train cytotoxicity classification models based on metal and ligand composition data.
• Screen ruthenium and iridium complexes for anticancer potential in a high-throughput computational pipeline.
• Predict cytotoxicity for metals with scarce experimental data using multi-metal models.
• Explore structure-activity relationships for transition-metal complexes via an AI-assisted web application.

Strengths

• Contains 26,500 experimental IC50 values, a substantial scale for the domain.
• Manually curated from 1,921 scientific articles, suggesting a systematic collection process.
• Covers 7,050 unique metal complexes tested against 754 cell lines, indicating broad experimental coverage.

Limitations

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

Provenance

Source

figshare

Collection Method

Manually curated from 1,921 scientific articles.

Time Range

null

Freshness

Last updated 2026-04-13 09:29:56; freshness should be verified.

Geography

null