Fig. 4
From: A quantitative prediction method utilizing whole omics data for biosensing
Application of OmicSense to various types of omics data. (A) Prediction errors (mean absolute error; MAE) in OmicSense compared with those in other commonly used methods for action potential amplitude, age, and pH associated with single-cell transcriptome (n = 31), metabolome (n = 42), and microbiome (n = 49) data structures, respectively. Different colors indicate different prediction methods, and different letters indicate significant differences between group means tested using the Tukey’s honest significant difference test (P < 0.05). (B) Correlation between the observed and predicted values in the prediction using OmicSense. For all cases, models were constructed using two-thirds of the available data, and prediction performance was validated using the remaining data.
