Feature analysis of VAECox

[yangguang8112]

At first, we extracted the top nodes with the highest variance in each of the second and third hidden layers. Then we calculated Pearson’s correlation between the values of each hidden node and the expression of each gene across all patient samples in the BRCA dataset.

Hello, I just read your paper, it is an interesting job, but I have a question. In section 3.3 of the paper, use the value of the third hidden layer and the amount of gene expression to calculate the Pierce correlation coefficient. The dimensions of these two vectors seem to be different. How is this calculated?

Thanks a lot!