FOLH1 Ligand Similarity Scores from ChemFusion and Tanimoto Metrics

A combined dataset integrating similarity scores for FOLH1 ligands from two distinct methods. The dataset, created by Rui Zhang and last updated in May 2026, contains 6.6 MB of data in CSV format. It facilitates a comparative analysis between the deep learning-based ChemFusionSimilarity model and the traditional Tanimoto similarity metric.

Use Cases

• Benchmarking deep learning similarity models against traditional fingerprint-based methods based on the comparative analysis described.
• Analyzing the correlation between ChemFusionSimilarity and Tanimoto scores for a specific target protein's ligands.
• Selecting or validating ligand candidates for the FOLH1 protein based on integrated similarity metrics.
• Training or evaluating new molecular similarity algorithms using a dataset with pre-computed scores from two established methods.

Strengths

• Provides a direct comparison between a deep learning and a traditional similarity metric for the same ligand set.
• Dataset is 6.6 MB in size, making it manageable for quick analysis and prototyping.
• Released under the permissive CC-BY-4.0 license, allowing for broad reuse and modification.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale model training.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect bias inherent to the specific FOLH1 ligand selection and similarity calculation methods used.

Provenance

Source

figshare, author Rui Zhang.

Collection Method

Likely generated by computational analysis integrating outputs from two similarity calculation models.

Time Range

null

Freshness

Last updated 2026-05-22 17:42:52; freshness should be verified.

Geography

null