Modeling systems-level dynamics: Understanding without mechanistic explanation in integrative systems biology. (February 2015)

• Record Type: Journal Article
• Authors: MacLeod, Miles; Nersessian, Nancy J.
• Is Part Of: Studies in history and philosophy of science. Volume 49(2015)
• Page Start: 1
• Page End: 11
• Publication Date: 2015-02
• Subjects: Systems biology -- Mechanistic explanation -- Understanding -- Systems-level understanding -- Abstraction -- Complexity; Biology -- Periodicals; Natural history -- Periodicals; Biology -- Philosophy -- Periodicals; Medicine -- History -- Periodicals; Medicine -- Philosophy -- Periodicals; Bioethics -- Periodicals; Biologie -- Histoire -- Périodiques; Biologie -- Philosophie -- Périodiques; Sciences de la santé -- Histoire -- Périodiques; Sciences de la santé -- Philosophie -- Périodiques; Médecine -- Histoire -- Périodiques; Médecine -- Philosophie -- Périodiques; 570.1
• DOI: 10.1016/j.shpsc.2014.10.004
• Languages: English
• ISSNs: 1369-8486

Abstract

Abstract: In this paper we draw upon rich ethnographic data of two systems biology labs to explore the roles of explanation and understanding in large-scale systems modeling. We illustrate practices that depart from the goal of dynamic mechanistic explanation for the sake of more limited modeling goals. These processes use abstract mathematical formulations of bio-molecular interactions and data fitting techniques which we call top-down abstraction to trade away accurate mechanistic accounts of large-scale systems for specific information about aspects of those systems. We characterize these practices as pragmatic responses to the constraints many modelers of large-scale systems face, which in turn generate more limited pragmatic non-mechanistic forms of understanding of systems. These forms aim at knowledge of how to predict system responses in order to manipulate and control some aspects of them. We propose that this analysis of understanding provides a way to interpret what many systems biologists are aiming for in practice when they talk about the objective of a "systems-level understanding." Highlights: An ethnographic account of model building practices in systems biology. We show that dynamic mechanistic explanation doesn't always account for epistemic goals in the field. We show modelers use abstraction strategies that obscure rather than clarify mechanisms for the sake of tractability. We show that complexity and data uncertainty often constrains modelers to more pragmatic forms of understanding.