biostudies-literatureUnknownKalia ANIGMS NIH HHS<h4>Motivation</h4>A major challenge in metabolomics is annotation: assigning molecular structures to mass spectral fragmentation patterns. Despite recent advances in molecule-to-spectra and in spectra-to-molecular fingerprint prediction (FP), annotation rates remain low.<h4>Results</h4>We introduce in this paper a novel tool (JESTR) for annotation. Unlike prior approaches that <i>explicitly</i> construct molecular fingerprints or spectra, JESTR leverages the insight that molecules and their corresponding spectra are views of the same data and effectively embeds their representations in a joint space. Candidate structures are ranked based on cosine similarity between the embeddings of query spectrum and each candidate. We evaluate JESTR against mol-to-spec, spec-to-FP, and specmol matching annotation tools on four datasets. On average, for rank@[1-20], JESTR outperforms other tools by 55.5% - 302.6%. We further demonstrate the strong value of regularization with candidate molecules during training, boosting rank@1 performance by 5.72% across all datasets and enhancing the model's ability to discern between target and candidate molecules. When comparing JESTR's performance against that of publicly available pretrained models of SIRIUS and CFM-ID on appropriate subsets of MassSpecGym dataset, JESTR outperforms these tools by 31% and 238%, respectively. Through JESTR, we offer a novel promising avenue towards accurate annotation, therefore unlocking valuable insights into the metabolome.ArXivarXiv:2411.14464v3https://www.ebi.ac.uk/biostudies/studies/S-EPMC11601792biostudies-literatureJESTR: Joint Embedding Space Technique for Ranking Candidate Molecules for the Annotation of Untargeted Metabolomics Data.PMC11601792R35 GM148219Krishnan DKalia AChen YZHassoun SfalseJESTR: Joint Embedding Space Technique for Ranking Candidate Molecules for the Annotation of Untargeted Metabolomics Data.<h4>Motivation</h4>A major challenge in metabolomics is annotation: assigning molecular structures to mass spectral fragmentation patterns. Despite recent advances in molecule-to-spectra and in spectra-to-molecular fingerprint prediction (FP), annotation rates remain low.<h4>Results</h4>We introduce in this paper a novel tool (JESTR) for annotation. Unlike prior approaches that <i>explicitly</i> construct molecular fingerprints or spectra, JESTR leverages the insight that molecules and their corresponding spectra are views of the same data and effectively embeds their representations in a joint space. Candidate structures are ranked based on cosine similarity between the embeddings of query spectrum and each candidate. We evaluate JESTR against mol-to-spec, spec-to-FP, and specmol matching annotation tools on four datasets. On average, for rank@[1-20], JESTR outperforms other tools by 55.5% - 302.6%. We further demonstrate the strong value of regularization with candidate molecules during training, boosting rank@1 performance by 5.72% across all datasets and enhancing the model's ability to discern between target and candidate molecules. When comparing JESTR's performance against that of publicly available pretrained models of SIRIUS and CFM-ID on appropriate subsets of MassSpecGym dataset, JESTR outperforms these tools by 31% and 238%, respectively. Through JESTR, we offer a novel promising avenue towards accurate annotation, therefore unlocking valuable insights into the metabolome.2025-01-01T00:00:00Z2025 Jun2025-08-14T03:06:12.806Z2025-04-04T02:35:16.786ZS-EPMC1160179239606728