Experimental NMR Peaks for 533,595 Chemical Compounds

533,595 chemical compounds with experimental NMR (Nuclear Magnetic Resonance) peak sequences, including both H-NMR and C-NMR data. The dataset, created by SpectrumWorld, provides SMILES representations and molecular formulas for each compound. It was last updated on December 8, 2025.

Use Cases

• Train machine learning models to predict NMR spectra based on SMILES representations.
• Validate computational chemistry methods by comparing predicted spectra to experimental peak sequences.
• Analyze correlations between molecular structure (via SMILES) and observed NMR chemical shifts.
• Build searchable databases for matching unknown experimental spectra against known compounds.

Strengths

• Contains 533,595 compounds, which suggests a substantial collection for analysis.
• Includes both H-NMR and C-NMR experimental peak data, providing complementary structural information.
• Pairs spectral data with SMILES strings and molecular formulas, enabling multi-modal analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect source bias inherent to the OpenDataLab experimental spectra database.

Provenance

Source

OpenDataLab experimental spectra database

Collection Method

Extracted experimental NMR peak sequences

Freshness

Last updated 2025-12-08 09:38:15