A living cell is essentially a molecular digital computer that configures itself as part of the execution of its code. By understanding how cells direct the assembly of their molecules, we can find ways to build computer chips that can self-organize, evolve and adapt to a changing environment. In this paper, we consider a model of the gene regulatory network of living cells called random multiple-valued network (RMN). An RMN can be tuned so that its statistical features match the characteristics of living cells. We present a set of algorithms for redundancy removal, partitioning and computation of attractors in RMNs. We also discuss how RMNs can be used for implementing logic functions.