MS-GPT model advances de novo molecular structure elucidation from mass spectra

Researchers have developed MS-GPT, a novel molecule-language model designed for de novo structure elucidation from tandem mass spectrometry (MS/MS) data. Unlike previous methods that rely on reference libraries or predefined candidates, MS-GPT directly generates molecular structures from spectra. It recasts the problem as spectrum-induced posterior querying, conditioning a language model on fingerprints and formulas. The model achieves state-of-the-art performance on the NPLIB1 and MassSpecGym datasets, demonstrating improved accuracy in identifying molecular structures. AI

IMPACT This research introduces a new approach for molecular structure elucidation using language models, potentially improving analytical chemistry workflows.

RANK_REASON The cluster contains an academic paper detailing a new model and its performance on specific datasets. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• MassSpecGym
• MS-GPT
• NPLIB1
• tandem mass spectrometry

• paper
• model release
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →