Chemical Structure Elucidation from Mass Spectrometry by Matching Substructures
This addresses a bottleneck in analytical chemistry for identifying chemical threats, but it is incremental as it builds on existing data-driven methods.
The paper tackles the problem of identifying unknown chemical structures from mass spectra and chemical formulas, which can take days for experts, by using neural networks to predict substructures and rank candidates, achieving over 90% micro F1-score for substructure classification and finding the correct structure in the top 20 candidates for 88% and 71% of test cases in two compound classes.
Chemical structure elucidation is a serious bottleneck in analytical chemistry today. We address the problem of identifying an unknown chemical threat given its mass spectrum and its chemical formula, a task which might take well trained chemists several days to complete. Given a chemical formula, there could be over a million possible candidate structures. We take a data driven approach to rank these structures by using neural networks to predict the presence of substructures given the mass spectrum, and matching these substructures to the candidate structures. Empirically, we evaluate our approach on a data set of chemical agents built for unknown chemical threat identification. We show that our substructure classifiers can attain over 90% micro F1-score, and we can find the correct structure among the top 20 candidates in 88% and 71% of test cases for two compound classes.