Hi dear friends,

I’ve a question about the number of PCs from ordination of DEICODE distance matrix.

My DEICODE matrix had 482 samples. The ordination showed 4503 features. But there was only 3 PCs and the proportion of variation was summed to 100%.

Is it right? And how come?

Because for traditional PCA, the number of PCs is either the number of samples or the number of features, whichever the smaller is the upper bound.

Thanks and look forward to our discussion.

Best,

Yun