I will first explain the question of existence of maximum likelihood estimators in Gaussian graphical models using the language of convex algebraic geometry. Based on this framework, I will then present results on the existence of maximum likelihood estimators for the bipartite graph $K_{2,m}$.