The original quantitative structure-activity relationship (QSAR) formulation was proposed by Hansch and Fujita in the 1960's and, since then QSAR analysis has evolved as a mature science, due mainly to the advances that occurred in the past two decades in the fields of molecular modeling, data analysis algorithms, chemoinformatics, and the application of graph theory in chemistry. Moreover, it is also worthy of note the exponential progress that have occurred in software and hardware development. In this context, a myriad of QSAR methods exist; from the considered "classical" approaches (known as two-dimensional (2D) QSAR), to three-dimensional (3D) and multidimensional (nD) QSAR ones. A distinct QSAR approach has been recently proposed, the receptor-dependent-QSAR, where explicit information regarding the receptor structure (usually a protein) is extensively used during modeling process. Indeed, a limited, but growing number of receptor-dependent QSAR methods are reported in the literature. With no intention to be comprehensive, an overview of receptor-dependent QSAR methods will be discussed along with an in-depth examination of their applications in drug design, virtual screen, and ADMET modeling in silico.