Fig. 2: Media blend and cytokine composition optimization PBMC cultures. | Nature Communications

Fig. 2: Media blend and cytokine composition optimization PBMC cultures.

From: Accelerating cell culture media development using Bayesian optimization-based iterative experimental design

Fig. 2

A Workflow and study parameters to optimize composition of media to maximize viability and maintain homeostasis of PBMCs in cultures. B PBMC cell viability as a function of the optimized media blend compared to individual standard media (DMEM, RPMI, XVIVO, and AR5) represented using data from six biological replicate experiments in each case. Individual double-sided t-test was performed between the standard media and optimal media formulation. *** signify a p-value < 0.01 with the exact values reported in the Source Data.xlsx file. C Comparison of the number of experiments to execute the different strategies for designing experiments considering 8 different cytokines. D Change in total cell density before and after 3 days of PBMC culturing using different compositions of cytokines with the red arrow indicating the condition meeting the desired objective. E Change in cell density of subpopulations of lymphocytes before and after 3 days in culture under different cytokine compositions. F Composition of the cytokines tested in the 12 different experiments (Expts). Red box corresponds to the condition meeting the desired objective (Expt. 11). Source data are provided in Source Data.xlsx file. PBMCs peripheral blood mononuclear cells, BO Bayesian Optimization, OFAT One Factor At Time, BBD Box Behnken Design, CCD Central Composite Design, FF Full factorial, FracFact Fractional Factorial, Frac CCD Fractional CCD, IL Interlukin, BAFF B-cell activating factor.

Back to article page