Synthetic Living Machines and the Emergence of Purpose

Understanding and managing purpose is an essential, practical, daily aspect of reverse-engineering life in basic evolutionary biology (or exobiology), and of forward-engineering new outcomes in regenerative medicine and synthetic bioengineering. We complement philosophy with powerful empirical study of multi-scale biological purposeful systems. We pose falsifiable hypotheses, evaluated using well-accepted scientific metrics: does a purpose-driven approach provide better prediction and control capabilities for complex biological outcomes? Our domain is the ability of cells to cooperate toward specific anatomical outcomes and behavioral capabilities in entirely novel configurations - purpose implemented by swarm intelligence of cellular collectives. If salamander embryo cells are combined with frog embryo cells, will the resulting chimera make legs or not? No existing model makes a prediction. Despite the deluge of molecular data, we are very far from prediction and control of form and function, nor have a theory of the scaling of goal states during multicellularity and of the cooperation/competition of subsystems with goals at different levels of organization. We will establish a model system for the study of how biological collectives acquire, pursue, recognize, and modify their purpose: novel living organisms that comprise cells from various species, and analyze their anatomical, physiological, and functional goals. This frees the experimenter from facile reliance on “eons of evolutionary selection” as an explanation for goals. By creating entirely novel bodies, not subject to the many frozen accidents of selection, we unlock highly tractable ways to experimentally address the origins and embodiment of purpose. By learning to predict and modify behavioral and anatomical goal states in novel creatures, our computational and experimental outcomes will show how a science of purpose and evolutionary developmental genetics not only coexist, but productively synergize.