Links between uniform Aztec diamonds and random matrices are numerous in the literature. In particular \cite{johansson2006eigenvalues,Forrester} established that, under correct rescaling, the probability density function of a certain subclass of dominos converges to the GUE-corners (GUE minor) process in the large size limit. We are interested to see whether this result holds when we modify the probability measure on the space of configurations. In the first part, we look at the case of biased Aztec diamonds, where different weights are associated to vertical and horizontal dominos. In the second part, we examine the case of two-periodic weightings. In both situations, we observe the convergence to GUE-corners with a rescaling that depends on the weighting.