Both stochastic and PDE modeling approaches have been used and compared in various context in biology. Typically, stochastic models are easier to parameterize, can be used to integrate underlying biological phenomena, but hard to analyze mathematically and usually computational expensive. PDE models are amenable to mathematical analysis and computational methods are well-developed, but biological processes are usually lumped together in them and parameters harder to obtain. Therefore it is important to find connections of different approaches. In this note, I give some investigation of methods to derive a Chapman-Kolmogorov type PDE model from a stochastic hybrid system, highlighting the close connection between these two.