Computes Moran's I along with expected value and standard deviation
under the null hypothesis. Optimized with Armadillo vectorization.
Usage
moranI_full_cpp(expr_matrix, W)
Arguments
- expr_matrix
Numeric matrix (genes x spots)
- W
Spatial weights matrix (spots x spots), row-standardized
Value
List containing observed I, expected I, and standard deviation
Details
This function pre-computes all weight matrix statistics once, then
loops over genes efficiently. The main bottleneck is the z * W * z'
computation which is O(n^2) per gene.