Figure 10.

Sample size for comparing protein abundance between conditions. Sample size for comparing protein abundance between conditions for a representative protein, in a future experiment with the same experimental structure. (a) Study of breast cancer cell lines. (b) Study of subjects with osteosarcoma. X-axis: minimal fold change that we'd like to detect. Y-axis: the number of biological replicates per group. Solid black line: the proposed joint modeling of all conditions. Dashed red line: a linear mixed effects model that represents the conditions of interest, and ignores the rest of the data. Dashed green line: a naïve analysis, which averages feature intensities in each replicate and uses a t-test for significance analysis. The anticipated proportion of differentially abundant proteins, the number of conditions and the number of technical replicates are as in the current experiments. The number of features per protein and the error variance are set to the median values.

Clough et al. BMC Bioinformatics 2012 13(Suppl 16):S6   doi:10.1186/1471-2105-13-S16-S6