Alzheimer’s disease (AD) is a progressive, degenerative neurological disorder for which there is no cure. The causative agent is ?-amyloid (A?) which becomes neurotoxic upon conformational change from ?-helix to ?-sheet. In silico methods have been used to indentify endogenous small molecules of the brain that are capable of binding to A? to inhibit conformational changes; this is a novel approach to the disease. Through the use of computational methods, several small molecules that are endogenous to the brain, such as phosphoserine, have been identified as being capable of binding to the monomeric forms of A?; in vitro studies support their role as anti-aggregants. One of the small molecules identified through these in silico methods, 3-hydroxyanthranilic acid (3HAA) has been developed through the use of Quantitative Structure-Activity Relationship (QSAR) studies to design more potent analogues. These in silico studies have also examined the capacity of synthetic compounds (structurally similar to endogenous molecules) to bind to both A? and other proteins affiliated with AD. Results indicate the potential for a single molecule to bind “promiscuously” to multiple proteins bearing a common BBXB (where B is a basic amino acid) motif affiliated with AD. This will allow for the development of molecules to target AD in a multifaceted approach. Furthermore, these small molecules can be selected through the use of “physinformatics” to bind with equal efficacy to the HHQK and LVFF regions (which play a role in the misfolding process) of A?; this will prevent conformational changes of the protein. A novel diagnostic imaging agent for AD has also been developed through computational methods; solapsone (formerly used to treat leprosy) has been identified as being structurally similar to species that bind to A? to initiate conformational changes. Results show that solapsone can chelate gadolinium, used in MRI, and bind to the soluble forms of A?, allowing for imaging of the toxic species in the human brain, and thus a definitive diagnosis of AD (which is not currently possible with living patients). Computational methods have proved useful in developing a new approach to treating AD, and designing a novel imaging agent.