From: A network-assisted co-clustering algorithm to discover cancer subtypes based on gene expression
Notation | Description |
---|---|
n | Number of samples |
d | Number of genes |
c | Number of sample clusters |
m | Number of gene clusters |
X | Gene expression matrix of size d × n |
X ∙i | The ith column of X, representing the expression of the ith sample |
X i∙ | The ith row of X, representing the expression of the ith gene |
F | Sample partition matrix of size n × c; F ij ∈[0,1]: F ij = 1 if X ∙i belongs to sample cluster j and F ij = 0 otherwise |
G | Feature partition matrix of size d × m; G ij ∈[0,1]: G ij = 1 if X i∙ belongs to gene cluster j and G ij = 0 otherwise |
S | A m × c matrix |
W | A d × d diagonal matrix; entries are the weights of the genes |