Skip to main content
. 2023 May 30;9:20. doi: 10.1038/s41540-023-00284-7

Fig. 3. Regular mode of active learning for medium optimization.

Fig. 3

A Boxplots of A450 of the cell culture at 168 h. Every 18–19 medium combinations predicted by the GBDT model using the culture data at 168 h are tested. Mean values for biological replicates (N = 3) are shown. The gray horizontal line indicates the cell culture with EMEM. An arbitrary unit is shown as a.u. B Boxplots of the prediction accuracy of the GBDT model built in each round. Mean absolute error (MAE), representing the average of the absolute difference between predicted and measured A450 values, was calculated by fivefold nested cross-validation. “All” indicates the accuracy evaluated using the entire dataset from the initial to Round 4. Asterisks indicate statistical significance by Mann‒Whitney’s U-test (p < 0.05). The boxplots show the median (center line), interquartile range (bounds of box), the range of typical data values (whiskers), and outliers (circle).