Look for condition covariates that correlate with condition partitions

The scoreColumnLabelingsForPCN attempts to find which column labeling best correlates with this principal component, alas in the case of this dataset there is only one labeling, and so the result any particular call not that interesting.

pcaginsu.scoreColumnLabelingsForPCN(1)

However one could determine which principal component best correlates with time, by looking at all of the principal components

for i in range(1,pcaginsu.rowPCAView.numCols+1):
  print i, pcaginsu.scoreColumnLabelingForPCN(i, verbose=False)[1]

Results in the following output.

1 [0.67150984942408032]
2 [0.69151881316901098]
3 [0.67958446890131774]
4 [0.67696146850058647]
5 [0.60690986817697534]
6 [0.63288334573057559]
7 [0.61963583669298239]
8 [0.60206151939136532]
9 [0.60206151939136521]
10 [0.56392206191915917]
11 [0.60206151939136532]
12 [0.57831316730290838]
13 [0.60206151939136532]
14 [0.0]
15 [0.60206151939136476]
16 [0.0]
17 [0.56392206191915917]

Apparently principal component 2 best explains time in this dataset.