The problem of non-fragile stabilization for the stochastic system with Markovian switching under partly unknown transition rates and actuator saturation is considered . By employing the parameter-dependent Lyapunov methodology, the non-fragile state feedback controller is proposed to guarantee the stochastic stability of the resulting closed-loop saturated system. Based on the obtained results, sufficient conditions with LMI constraints are established to acquire the largest contraction invariant set in the mean square sense. Finally, an example is given to illustrate the effectiveness of the proposed method.