Skip to main content

Table 2 Predictive modeling of nutrition condition of single cells

From: Quantitative dynamics of triacylglycerol accumulation in microalgae populations at single-cell resolution revealed by Raman microspectroscopy

Fingerprint region
  6 h 12 h 24 h 36 h 48 h 72 h 96 h*
Number of PCs used in LDA 9 9 2 2 2 2 2
Variations explained by these PCs (%) 67.5 82.0 72.6 89.1 91.1 85.7 93.6
MCR of calibration data (%) 2.2 2.2 1.1 1.1 1.1 1.1 0
MCR of validation data (%) 6.7 3.3 3.3 0 0 0 0
Hydrocarbon region
  6 h 12 h 24 h 36 h 48 h 72 h 96 h*
Number of PCs used in LDA 10 12 5 2 2 2 2
Variations explained by these PCs (%) 89.1 95.8 94.9 96.3 94.7 97.9 97.2
MCR of calibration data (%) 2.2 0 0 0 0 0 0
MCR of validation data (%) 16.7 6.7 0 0 0 0 0
  1. *For both calculations, one outlier was excluded from the training set at 96 h.
  2. PC-LDA results were shown for discriminating cells between Groups N- and N + at each time point. h: hours; LDA: linear discriminant analysis; MCR: misclassification rate by leave-one-out cross validation (LOOCV); PC: principal component.