In this study, a rigid robot is hybridized with a soft part to improve its properties of relative positioning and safe interaction. While the design of the device is quite simple, the kinematics, on the other hand, becomes complex. Pre-operative planning of such a robot is a challenge that is faced in this paper. The clinical context is the automation of transrectal prostate biopsy: a robot is connected to a thin ultrasound (US) probe instrumented with a needle through a flexible silicone part. The procedure consists of: moving the probe near the prostate through the rectum, aligning the probe with identified lesions in the prostate, and shooting the needle to collect tissue samples. In this paper we demonstrate in simulation the feasibility of our methods to solve the planning of the entire procedure. The approach we propose is based on numerical models of both robot and soft tissues, using the Finite Element Method (FEM) to model the soft parts, and a mapping mechanism to propagate the forces and displacements between the soft and rigid models. Finally, we extend the optimization algorithm proposed in previous work for the relative positioning of the hybrid robot on targets in the prostate, as it interacts with the soft tissues. With this formulation we can also set an upper limit to the force the robot is applying on its surroundings, which is particularly critical for the application we study in this paper.