The paper focuses on the modeling and simulation of heterogeneous bio-compatible systems using SystemC and its AMS extension SystemC-AMS. The studied lab-on-chip takes as input a DNA sample, represented by its initial chemical concentration and the sequence of DNA characters representing the DNA sequence. The sample is amplified by Polymerase Chain Reaction, separated by Capillary Electrophoresis and optically detected by fluorescence. A manycore digital architecture controls the analog systems on one hand, and executes multithreaded complex DNA alignments on the other. The analog parts are modeled in SystemC-AMS, using the Timed Data Flow Model of Computation (TDF MoC). The digital part is modeled with systemC and optimized components and models dedicated to the simulation of manycore architectures. First, The paper discusses the modeling and optimization of each part independently from the others, then it details how these parts can be interconnected to simulate the whole system. Results are shown, and special attention is given to the simulation timesteps used to overcome the huge time scale difference between the digital and analog parts. Finally, the paper proposes potential solutions that could be the next step towards the modeling, simulation, and the refinement of complex heterogeneous systems.