Process-based Modelling of RNA and Protein Interactions: a Formal Approach

Process algebras and agent-based models have proven to be effective methods for studying biologi- cal systems. Our research employs such techniques to investigate the behaviours that characterise biological macromolecules and reveal the global properties of biochemical processes resulting from local molecular interactions. This dissertation consists of two parts. In the first one, we use formal methods, such as the Calculus of Communicating Systems, to demonstrate the existence of a con- gruence level at which the folding of RNA molecules is behaviourally equivalent to that of proteins. This finding allows us to hypothesise the role that RNA functional complexity played during the evo- lutionary process that led proteins to emerge as the primary catalysts in modern cells. We also rely on such a representation to model how an error in the genetic code–i.e., a mutation–can propagate through each step of the synthesis of a new protein, ultimately affecting its folded conformation. We formally prove that the different complexity of RNA and protein folding results in significantly dis- similar impacts that a single nucleotide mutation can have on the structures of proteins compared to those of RNAs. In the second part of this manuscript, we describe an agent-based approach that we specially designed to investigate the global behaviour of long-distance electrodynamic interac- tions among biomolecules. Agents are software entities that can perceive their environment and operate on it autonomously. Using agent-oriented programming, we created a software replica of glycolysis–the metabolic process that provides energy to cells through glucose oxidation. The abil- ity of agents to reproduce molecular behaviours makes it possible to study biochemical processes in a virtual environment and interpret them as the result of underlying molecular interactions. Fur- thermore, the generated agent interaction matrix can be filtered using topological data analysis, allowing us to investigate the role of 2-simplex formation in biochemical reactions. Our goal is to understand how specific types of molecular interactions influence glycolysis effectiveness, par- ticularly in cancer cells. The two parts that make up our work represent the main phases of the engineering life cycle for the simulation of enzyme behaviour; they are intended as the preliminary steps in the development of a computational framework able to contribute to cancer studies per- formed on experimental data. This research sheds new light on how biomolecules interact and lays the groundwork for in silico personalised and precision medicine.