We have developed a new strategy for identifying bacterial species in biological samples using specific LC-MS/MS peptidic signatures. In the first training step, deep proteome coverage of bacteria of interest is obtained in Data Independent Acquisition (DIA) mode, followed by the use of machine learning to define the peptides the most susceptible to distinguish each bacterial species from the others. Then, in the second step, this peptidic signature is monitored in biological samples using targeted proteomics. This method, which allows the bacterial identification from clinical specimens in less than 4h, has been applied to 15 species representing 84% of all Urinary Tract Infections (UTI).
This dataset contains all the DDA files used to create bacterial spectral libraries prior to DIA analyses.
[dataset license: CC0 1.0 Universal (CC0 1.0)]
Keywords: Bacterial identification ; Urine ; LC-MSMS ; DIA ; Machine Learning
|Principal Investigators:||Arnaud Droit, CHU de Quebec Universite Laval, Canada|
Roux-Dalvai F., Gotti C., Leclercq M., Hélie MC., Boissinot M., Arrey T.N., Dauly C., Fournier F., Kelly I., Marcoux J., Bestman-Smith J., Bergeron M.G., Droit A.
Fast and accurate bacterial species identification in urine specimens using LC-MS/MS mass spectrometry and machine learning.
Mol Cell Proteomics. 2019 Oct 4. pii: mcp.TIR119.001559. doi: 10.1074/mcp.TIR119.001559.
|Number of Files:|
Protein Accessions (reported):
FTP Download Link (click to copy):