Rapid identification of lactic acid bacteria at species/subspecies level via ensemble learning of Ramanomes

• DOI: 10.3389/fmicb.2024.1361180
• Publication Date: 2024-04-08T05:00:39Z
• Authors: Yan Ren; Yang Zheng; Xiaojing Wang; Shuang Qu; Lijun Sun; Chenyong Song; Jia Ding; Yuetong Ji; Guoze Wang; Pengfei Zhu; Likun Cheng
• Keywords: Subspecies; Identification

Rapid and accurate identification of lactic acid bacteria (LAB) species would greatly improve the screening rate for functional LAB. Although many conventional molecular methods have proven efficient reliable, LAB using these has generally been slow tedious. Single-cell Raman spectroscopy (SCRS) provides phenotypic profile a single cell can be performed by (which directly detects vibrations chemical bonds through inelastic scattering laser light) an individual live cell. Recently, owing to its affordability, non-invasiveness, label-free features, Ramanome emerged as potential technique fast bacterial detection. Here, we established reference database consisting SCRS data from 1,650 cells nine species/subspecies conducted further analysis machine learning approaches, which high efficiency accuracy. We chose ensemble meta-classifier (EMC), is suitable solving multi-classification problems, perform in-depth mining data. To optimize accuracy algorithm, compared classifiers: LDA, SVM, RF, XGBoost, KNN, PLS-DA, CNN, LSTM, EMC. EMC achieved highest average prediction 97.3% recognizing at level. In summary, Ramanomes, with integration EMC, promising in laboratories may thus developed sharpened direct fermented food.