Planetary systems when examined as a single unit of a physical system are more than the sum of its parts i.e. host star(s), planets, minor bodies, etc. In this thesis, I propose several system-level properties of planetary systems and explore their implications. The system-level analysis approach provides us with many unsolicited benefits: novel ground for comparing theory with observations, new predictions for testing planet formation models, and a different perspective on what it means for an exoplanetary system to be similar to the Solar System (or Solar System analogue). In addition, the frameworks and concepts introduced here provide us with new ways to quantify, and characterize individual planetary systems and compare one system with another. Based on these novel system-level parameter spaces, I propose and analyse two different approaches to classifying planetary systems.

Using synthetic populations from the Bern Model and a new computer pro- gram, KOBE to simulate the transit survey, it is shown that the peas in a pod trend in the architectures of exoplanetary systems are highly likely to be of astro- physical origin (as opposed to emerging from detection biases). I introduce a novel model-independent framework for the architecture of exoplanetary systems. The space of system architecture is partitioned into four architecture classes: Similar, Mixed, Ordered, & Anti-Ordered. Among several findings, it is shown that Sim- ilar architectures are the most common outcome of core-accretion based planet formation and most planets in this architecture class usually form in-situ. Inves- tigation of the formation pathways of the four architecture classes results in the prediction of a new metallicity-architecture correlation. It is shown that initial conditions (or nature) play an essential role for Similar class, while dynamical interactions (or nurture) govern the emergence of other architecture classes. The concept of the state of a planetary system is introduced. A diagram depicting the state of thousands of synthetic and observed planetary systems shows unexpected structure, and allows the evolution of planetary systems to be systematically stud- ied. Using the mismatch between theoretical and observed states, the physical characteristics of undetected exoplanets is predicted.