Representing a new class of tool for biological modeling, Bio Model Analyzer (BMA) uses sophisticated computational techniques to determine stabilization in cellular networks. This paper presents designs aimed at easing the problems that can arise when such techniques -distinct approaches to conceptualizing networks – are applied in biology. The work also engages with more fundamental issues being discussed in the philosophy of science and science studies. It shows how scientific ways of knowing are constituted in routine interactions with tools like BMA, where the emphasis is on the practical business at hand, even when seemingly deep conceptual problems exist. For design, this perspective refigures the frictions raised when computation is used to model biology. Rather than obstacles, they can be seen as opportunities for opening up different ways of knowing.