Abstract: The objective of this dissertation was to design small molecule drug candidates for different disease targets by understanding the energetics and dynamics of their binding protein/enzyme/receptor partners. Protein-protein interactions are intrinsic to virtually every cellular process such as transcription regulation and signal transduction, and inappropriate protein-protein interactions may lead to human diseases such as cancer. These interactions commonly rely on a few key residues ("hot spot residues") and single point mutations of "hot spot" residues could cause disruption of theses protein complexes. Hence, small molecule antagonists, which interfere mainly with critical amino acid contacts, could have significant outcomes on disruption of binding equilibrium of protein/protein complex. By utilizing this concept, we have designed IL-6 inhibitors to disrupt interactions between IL-6 and gp130 (chapter 2, 3 and 4). Traditional drug discovery begins by identifying the protein target related to disease and finding a lead compound, a potential drug that bears the desired physical and biological features from a library of known chemical compounds. This limits the search space from the beginning and makes new drug discovery (new chemical structure) a very difficult task. However, as the cellular and molecular mechanisms behind many diseases are increasingly understood, new avenues for rational drug development emerge. This can be complemented by structure based drug design methods that utilize three dimensional structure of the target protein. Recent advancements in computational techniques and hardware have helped researchers using in silico methods to a speedy lead identification and optimization. Large virtual chemical libraries are now available for screenings that lead to discovery of small molecule inhibitors of HIV-IN and LEDGF interactions (Chapter 5 and 6). Protein/receptor structures are not static in the body; they often bear plasticity by accommodating chemically diverse ligands. Also multiple receptor conformation exists in their dynamic equilibrium. Thus, a single conformation is not enough to understand the activation mechanism of the receptor. We have utilized molecular dynamics simulations methods to obtain ensemble receptor conformations (Chapter 7). These ensemble conformations may represent different conformational state on energy landscape. Active like state obtained from molecular dynamics simulation was used to create three dimensional models of proteins with unknown structures, i.e. adrenergic receptors (Chapter 8). The main purpose of this thesis work was to understand underlying interaction between protein binding partners and design high affinity small molecules by computational techniques. Small molecules by mimicking "hot spots" residues or by binding to "hot spot" residues of target protein were designed to disrupt protein-protein interactions. With the work reported in this thesis, we aim to contribute to the field of computational drug discovery. We have attempted to estimate the ligand affinities to a protein structure by simulating the formation of protein-ligand complexes. Further, in this thesis, we will show that our computational approach helped to design diverse small molecules for different disease targets.