molclaw-mol-similarity

What problem does it solve?

Rapidly quantify how similar candidate small molecules are to a target molecule by computing Tanimoto similarity scores and counting shared structural fragments using Morgan fingerprints, enabling data-driven prioritization in screening and hit triage.

Core Features & Use Cases

• Tanimoto similarity calculation between a target SMILES and a list of candidate SMILES using configurable Morgan fingerprint radius and bit size.
• Shared fragment counting that returns the number of common structural fragments per candidate to support scaffold analysis and diversity filtering.
• Use Cases: prioritize virtual screening hits by similarity, cluster compounds for lead selection, or compute similarity metrics for downstream ML models.

Quick Start

Compare the target SMILES CC(=O)Oc1ccccc1C(=O)O against a list of candidate SMILES to compute Tanimoto similarity scores and shared fragment counts with a Morgan radius of 2.