Advanced Photonics, Volume. 7, Issue 4, 046007(2025)
Bacterial identification by metabolite-level interpretable surface-enhanced Raman spectroscopy
Fig. 1. SERSome acquisition of bacteria lysate. (a) Schematic workflow including thermal lysis of bacteria, removal of debris, addition of Ag NPs and salt ions, injection of the mixture into a capillary tube, and detection using Raman spectroscopy. (b) Extinction spectrum and TEM image of Ag NPs. (c) Heatmaps of SERSomes and (d) the typical single SERS spectra from
Fig. 2. Analysis of SERS spectra of eight bacteria. (a) Average SERS spectra of eight bacteria. The 200 to
Fig. 3. Bacteria identification using interpretable CNN. (a) Architecture of CNN for bacterial classification. SHAP is used for model interpretation. (b) Confusion matrix on the testing set. (c) SHAP values of eight bacteria species. For clarity, an average spectrum overlaid with SHAP values is also shown at the top.
Fig. 4. Metabolite identification by Ag NP-assisted LDI-MS. (a) Workflow of Ag NP-assisted LDI-MS. The Ag NP-adsorbed metabolites were separated and dropped on the LDI plate and measured by LDI-MS. (b) Spearman’s correlations among samples (
Fig. 5. Metabolite-level interpretation of bacterial SERS spectra. (a) Schematic workflow for spectral decomposition. The metabolite panel was established based on the results of LDI-MS. Coefficients were derived by spectral decomposition based on NNLS. (b) Representative original bacterial spectra (black) and fitting spectra (blue) of the eight bacteria. (c) Box plots based on cosine similarity between the original spectrum and the fitting spectrum. Confusion matrices of (d) random forest and (e) support vector machine for bacterial identification based on coefficients from spectral decomposition.
|
|
Get Citation
Copy Citation Text
Haoran Chen, Ruike Zhao, Xinyuan Bi, Nan Shen, Xi Mo, Yue Tao, Zhou Chen, Jian Ye, "Bacterial identification by metabolite-level interpretable surface-enhanced Raman spectroscopy," Adv. Photon. 7, 046007 (2025)
Category: Research Articles
Received: Mar. 14, 2025
Accepted: May. 26, 2025
Published Online: Jun. 23, 2025
The Author Email: Yue Tao (taoyue@scmc.com.cn), Zhou Chen (chenzhou96@sjtu.edu.cn), Jian Ye (yejian78@sjtu.edu.cn)