Hello all, I am trying to cluster the PCA analysis output (PC loadings). The rough version of protocol (attached) is working but i need help in:
'how to label the properties in the PC clustering visualization'. The outcome i am currently getting is this (shown in figure below):
And the data I am reading is:
Properties | PC1 | PC2 |
PC_ | 0.21297 | -0.096328 |
Q_PC_ | 0.21297 | -0.096328 |
vsa_pol | -0.20656 | -0.029011 |
lip_acc | -0.21372 | 0.016597 |
PEOE_PC_ | 0.21264 | -0.074102 |
TPSA | -0.20802 | 0.034297 |
PEOE_VSA_PNEG | -0.19425 | -0.15775 |
Q_VSA_PNEG | -0.18566 | -0.22706 |
GCUT_SLOGP_1 | 0.16598 | 0.23982 |
Q_VSA_POL | -0.20697 | 0.083965 |
SlogP_VSA2 | -0.18747 | 0.038381 |
SMR_VSA0 | -0.1734 | -0.24447 |
vsa_acc | -0.16569 | -0.31525 |
Q_RPC_ | 0.18805 | 0.0052587 |
RPC_ | 0.18805 | 0.0052587 |
a_ICM | -0.141 | -0.37005 |
Q_VSA_FHYD | 0.1739 | 0.17617 |
Q_VSA_FPOL | -0.1739 | -0.17617 |
a_nO | -0.19662 | 0.040441 |
PEOE_VSA_POL | -0.19927 | 0.10687 |
a_nN | -0.16072 | -0.021941 |
PEOE_VSA_6 | -0.16123 | -0.055919 |
vsa_other | -0.16851 | -0.23518 |
a_IC | -0.1812 | 0.28785 |
SlogP | 0.16429 | -0.04757 |
PC. | -0.18259 | 0.31464 |
Q_PC. | -0.18259 | 0.31464 |
zagreb | -0.1707 | 0.25108 |
Weight | -0.17853 | 0.23416 |
One more question I have is how to delete rows in a particular outcome? For instance I am running a PCA protocol whose output in PCA loadings is Std. Dev (1st row), Variance explained (2nd row) and Total variance explained (3rd row), after which (4th row and further) all the properties start.
When i pipe this output in another component (such as above mentioned 'clusering PCA visualization') it automatically reads the first three rows also (where i only want it to read the properties). How do i avoid reading such rows?
Thank you
nar