Power calculation for microarray data
0
0
Entering edit mode
10 months ago
Gene_MMP8 ▴ 210

I have an initial sample of 228 patients from a microarray study. Recently I have obtained a new set of labels specifying different condition types for only 69 out of the 228 patients. I wanted to run a DEG analysis on this set of 69 patients in order to identify potential biomarkers.

However, while doing the analysis using limma, I didn't get any significant results (adj p-value =1 for all genes). A small subset of the results is shown below. Is it possible to do some sort of power analysis to show that we have very little signal in this data owing to the small sample size?

     ID  logFC AveExpr      t P.Value adj.P.Val     B
7415  IFI44L -0.875    7.19 -1.308   0.195         1 -4.62
18598 TUBB2A -0.819    7.06 -1.874   0.065         1 -3.99
7418   IFIT1 -0.722    7.33 -1.215   0.229         1 -4.70
7410   IFI27 -0.619    8.00 -0.972   0.335         1 -4.89
15226  RSAD2 -0.590    7.74 -1.124   0.265         1 -4.77
5653   FGF13  0.549    6.18  1.387   0.170         1 -4.54

statistical_power microarray • 226 views