SERS-assisted characterization of cell biomass from biofilm-forming <i>Acinetobacter baumannii</i> strains using chemometric tools
Arslan Yousaf, Muhammad Hafeez Ullah, Haq Nawaz, Muhammad Irfan Majeed, Nosheen Rashid, Abdulrahman Alshammari, Norah A. Albekairi, Arslan Ali, Munawar Hussain, Abu Bakar Salfi, Muhammad Aamir Aslam, Kinza Idrees, Allah Ditta
Abstract
. Chemometric tools, such as principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA), were employed for the classification and differentiation of SERS spectra from bacterial strains with varying biofilm-producing capacities, achieving 100% sensitivity, 94.3% specificity, and an area under the curve (AUC) value of 0.81 through Monte Carlo cross-validation. Furthermore, K-fold (Leave-K-out cross-validation (LKOCV)) was applied to verify the robustness of the PLS-DA model, and the AUC value was found to be 0.90, with a sensitivity of 100% and specificity of 98%. These results demonstrate that the PLS-DA model is highly effective for the differentiation and classification of bacterial strains with varying capacities for biofilm production.