This thesis discusses our efforts in using the energy landscape theory and coarse-grained molecular dynamics protein folding models to explore the folding energy landscape of proteins. The Associative-memory, Water-mediated, Structure and Energy Model (AWSEM) is capable of performing de novo structure prediction on not only many natural globular proteins but also designed proteins such as Top7 and Takada. AWSEM also enables us to investigate the robustness of folding natural and designed protein sequences upon simplification of full sequences to the five-letter or two-letter code. More recent work, using AWSEM or structure-based (SB) model with the addition of an implicit membrane energy term, shows that the energy landscapes for folding α-helical membrane proteins are funneled once their native topology within the membrane is established, further proves that tertiary folding of α-helical membrane proteins is thermodynamically controlled. The first chapter is an overview of the energy landscape theory of protein folding, followed by subsequent three chapters which describe in details how the energy landscape theory can be used as a fundamental theoretical framework to elucidate the folding problems (folding and binding) for both globular (natural and designed) proteins and α-helical membrane proteins.