Hypothesis-and-Refinement Learning of Organic Structures from Multimodal Spectroscopic Data
Merged summary
TL;DR - SpectroMol combines multimodal NMR evidence with a mass-constrained molecular generator to automate organic structure elucidation. It achieves 93.8% top-1 accuracy on simulated data and transfers to experimental spectra with limited fine-tuning.
- QM9SPIN provides DFT-derived 1D and 2D spectra, including J-coupling, DEPT, and explicit spin interactions.
- SpectroMol proposes chemically valid structures from multimodal spectral inputs.
- MS-Mol2Mol applies molecular formula, exact mass, and unsaturation constraints using a prior trained on 400 million molecules.
- Mass-guided refinement further improves predictions on experimental spectra.
Sources (1)
Hypothesis-and-Refinement Learning of Organic Structures from Multimodal Spectroscopic Data
TL;DR - SpectroMol combines multimodal NMR evidence with a mass-constrained molecular generator to automate organic structure elucidation. It achieves 93.8% top-1 accuracy on simulated data and transfers to experimental spectra with limited fine-tuning.
- QM9SPIN provides DFT-derived 1D and 2D spectra, including J-coupling, DEPT, and explicit spin interactions.
- SpectroMol proposes chemically valid structures from multimodal spectral inputs.
- MS-Mol2Mol applies molecular formula, exact mass, and unsaturation constraints using a prior trained on 400 million molecules.
- Mass-guided refinement further improves predictions on experimental spectra.