Machine Learning-Assisted False Positive Detection in Metabolite Identification Workflows
Ramon Adàlia, Paula Cifuentes, Joyce Liu, Lionel Cheruzel, Gemma Sanjuan, Tomàs Margalef, Ismael Zamora · 2025
Read Machine Learning-Assisted False Positive Detection in Metabolite Identification Workflows by Ramon Adàlia, Paula Cifuentes, Joyce Liu, Lionel Cheruzel, Gemma Sanjuan, Tomàs Margalef, Ismael Zamora (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12750405.txt.
Description
Ramon Adàlia, Paula Cifuentes, and their co-authors built a machine learning tool to catch false hits when labs identify drug breakdown products in the body. Their 2025 study notes that the data from liquid based separation and mass based measures are complex, so many peaks look like real products but are not. Sorting them by hand slows drug work.
The team built a feature set from expert knowledge, covering mass spectra, peak shapes, and how signals change over time. They trained tree based classifiers on public data and on real company data, covering both small drugs and newer drug types. The tool cut false hits by a clear margin. Fewer false leads mean labs can spend their time on the products that are real.
Catalog record
| CREATOR | Ramon Adàlia, Paula Cifuentes, Joyce Liu, Lionel Cheruzel, Gemma Sanjuan, Tomàs Margalef, Ismael Zamora |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |