We utilize a new method, bottom-up building block assembly, to efficiently generate energy- minimizing structures for a family of complex, interacting patchy particles for which a priori knowledge of preferred assembled structures is lacking. We introduce the concept of self-assembly propensity and show it depends on the assembly pathways available to the building blocks. By grouping generated configurations into motifs we discern important characteristics for assembling specific patterns and identify building blocks with a propensity to self-assemble a desired configuration. We utilize the assembly pathways to propose strategies for overcoming poor self-assembly propensity.