2026-10-10T03:52:16Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33950461 2026-09-21T06:07:31Z category_4 category_146 category_7 category_8 category_12 category_21 category_873 category_734 category_69 category_132 portal_63 item_type_3 month_year_09_2026

Rapid Species-Level Classification of Urinary Pathogens from Raw LC-MS/MS Signals Using Machine Learning Simon J. Pelletier (25080802) Antoine Lacombe-Rastoll (25080805) Florence Roux-Dalvai (344244) Mickaël Leclercq (11377902) Clarisse Gotti (11377893) Eve Bérubé (25080808) Pascaline Bories (25080811) Marie-Ève Thibeault (25080814) Dorte B. Bekker-Jensen (4694626) Nicolai Bache (1677517) Maciej Bromirski (1592047) Sandra Isabel (10052952) Frederic Precioso (9419527) Arnaud Droit (36418) Biochemistry Space Science Medicine Microbiology Cell Biology Biotechnology Chemical Sciences not elsewhere classified Biological Sciences not elsewhere classified Inorganic Chemistry Infectious Diseases traditional interpretation stack tof mass spectrometry method achieved high diagnosis still depends conventional pipelines depend consuming workflows based approach reached 91 protein identification steps clinical mass spectrometry 0 false positives level microbial identification fast microbial identification direct microbial diagnosis 5 </ sup level classification fast enough clinical settings 5 min microbial culture urinary pathogens specificity needed results show raw lc promising direction produces species preserving information machine learning lengthy analyses enabled raw confidence classification concept demonstrates common infections analytical framework although lc 86 ). Urinary tract infections are among the most common infections in humans, yet their diagnosis still depends on time-consuming workflows based on microbial culture, followed by MALDI-TOF mass spectrometry. Although LC-MS/MS offers the sensitivity and specificity needed to bypass culture, conventional pipelines depend on lengthy analyses and peptide/protein identification steps, limiting the throughput and hindering its adoption in clinical settings. Here, we introduce a direct, identification-free LC-MS/MS workflow that analyzes raw ion signal and produces species-level microbial identification in about 5 min after preparation, fast enough to meet clinical throughput requirements. Our machine learning-enabled raw-signal pipeline bypasses peptide identification entirely, preserving information and eliminating the traditional interpretation stack. Across 15 independent analytical batches covering 28 clinically relevant pathogens, the method achieved high-confidence classification (MCC = 0.86). Applied to 206 clinical urine specimens across three batches, the approach reached 91% accuracy at clinically actionable microbial loads (greater than 10⁵ CFU/mL) and, critically, 0 false positives in control specimens. The performance was lower for specimens below this threshold. These results show that raw LC-MS/MS spectra contain sufficient biological information for direct microbial diagnosis, establishing an analytical framework for clinical mass spectrometry. This proof-of-concept demonstrates that rapid, culture-free, fast microbial identification is achievable and positions raw signal inference as a promising direction for next-generation diagnostic mass spectrometry. 2026-09-21T00:00:00Z Dataset Dataset 10.1021/acs.analchem.6c02725.s004 https://figshare.com/articles/dataset/Rapid_Species-Level_Classification_of_Urinary_Pathogens_from_Raw_LC-MS_MS_Signals_Using_Machine_Learning/33950461 CC BY-NC 4.0