Negative Data for Drug Discovery Pipeline Validation via Computational Chemistry

A dataset of negative data points for validating virtual high-throughput screening (VHTS) pipelines. The data was generated by Stefan M. Ivanov using computational methods to randomize ligands across experimental structures and create structural isomers of known binders. The dataset was last updated on March 18, 2026.

Use Cases

• Benchmarking molecular docking tools based on negative data matching positive binders in molecular properties.
• Validating multi-step VHTS pipelines to assess enrichment at each step based on the described validation method.
• Training or testing machine learning models for binding affinity prediction using the generated negative ligand sets.

Strengths

• Data is generated to match positive binders closely in key molecular properties, as described in the methodology.
• Dataset is licensed under CC-BY-4.0, allowing for open sharing and modification.
• The generation method requires no additional experimental work, offering a cost-effective validation resource.

Limitations

• The dataset is small at 18.4 KB, indicating limited scope.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

figshare, author Stefan M. Ivanov

Collection Method

Generated computationally by randomizing ligands across published experimental structures and generating structural isomers of known binders.

Freshness

Last updated 2026-03-18 07:37:36; freshness should be verified.