Search

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

ABSTRACT To reveal the dynamic features of cellular systems, such as the correlation among phenotypes, a time or condition series set of samples is typically required. Here, we propose intra-ramanome correlation analysis (IRCA) to achieve this goal from just one snapshot of an isogenic population, via pairwise correlation among the cells of the thousands of Raman peaks in single-cell Raman spectra (SCRS), i.e., by taking advantage of the intrinsic metabolic heterogeneity among individual cells. For example, IRCA of Chlamydomonas reinhardtii under nitrogen depletion revealed metabolite conversions at each time point plus their temporal dynamics, such as protein-to-starch conversion followed by starch-to-triacylglycerol (TAG) conversion, and conversion of membrane lipids to TAG. Such among-cell correlations in SCRS vanished when the starch-biosynthesis pathway was knocked out yet were fully restored by genetic complementation. Extension of IRCA to 64 microalgal, fungal, and bacterial ramanomes suggests the IRCA-derived metabolite conversion network as an intrinsic metabolic signature of isogenic cellular population that is reliable, species-resolved, and state-sensitive. The high-throughput, low cost, excellent scalability, and general extendibility of IRCA suggest its broad applications. IMPORTANCE Each isogenic population of cells is characterized by many phenotypes, which change with time and condition. Correlations among such phenotypes are fundamental to system function, yet revelation of such links typically requires multiple samples. Here, we showed that, by exploiting the intrinsic metabolic heterogeneity among individual cells, such interphenotype correlations can be unveiled via just one snapshot of an isogenic cellular population. Specifically, a network of potential metabolite conversions can be reconstructed using intra-ramanome correlation analysis (IRCA), by pairwise correlation of the thousands of Raman peaks or combination of peaks among single-cell Raman spectra sampled from just one instance of the cellular population. The ability to rapidly and noninvasively reveal intermetabolite conversions from just one snapshot of one sample should usher in many new opportunities in functional profiling of cellular systems.