Fig. 3From: Design of experiments to assess the effect of culture parameters on the osteogenic differentiation of human adipose stromal cellsCalculated percent contributions of the variables toward the ALP expression. The variables had either a net positive or a net negative contribution to the ALP expression. If the (+) variability of the variable resulted in higher ALP expression, the contribution was positive; if it led to lower ALP expression, the contribution was deemed negative. The average experimental error between the three biological replicates (4.1%) was used as the significance limit, which resulted in the following ranking of the variables in terms of their final contribution: (1) BMP-9, (2) serum, (3) hASC source, (4) AP, (5) DEX, and (6) seeding densityBack to article page