Researchers' Assumptions and Mathematical Models: A Philosophical Study of Metabolic Systems Biology

Abstract

This thesis examines the philosophical implications of the assumptions made by researchers involved in the development of mathematical models of metabolism. It does this through an analysis of several detailed historical case studies of models between the 1960’s and the present day, thus also contributing to the growing literature on the historiography of biochemical systems biology. The chapters focus on four main topics: the relationship between models and theory, temporal decomposition as a simplifying strategy for building models of complex metabolic systems, interactions between modellers and experimental biochemists, and the role of biochemical data. Four categories of assumptions are shown to play a significant role in these different aspects of model development; ontological assumptions, idealising assumptions, assumptions about data, and researchers’ commitments. Building on this analysis, the thesis brings to light the importance of researcher’s ontological and idealising assumptions about the temporal organisation, alongside the spatial organisation, of metabolic systems. It also offers an account of different forms of interactions between research groups – hostile interactions, closed collaboration, and open collaboration – on the basis of differences in the characteristics of researcher’s commitments. Throughout the case studies, biological data play a powerful role in model development by virtue of the contents of available data sets, as well as researchers’ perceptions of those data, which are in turn influenced by their ontological assumptions. The historical trajectories explored illustrate how the relationships between different facets of model building, and their associated philosophical abstractions, are often best understood as transient features within a highly dynamic research process, whose role depends on the specific stage of modelling in which they are enacted. This thesis provides an expanded perspective on the different types and roles of assumptions in the development of mathematical models of metabolism, which is firmly grounded in a historical analysis of scientific practice.